The Sleep Disorders

Home , Dyssomnia, Horlicks, Night terror, REM rebound, Second wind (sleep), Sleep induction

The Sleep Disorders

Peter Hauri, PhD

Associate Professor of Psychiatry (Clinical Psychology) Director, Dartmouth-Hitchcock Sleep Clinic Dartmouth Medical School Hanover, New Hampshire Contents

5 Introduction The Sleep Laboratory I 5 6 Basic Facts on Sleep Sleep Architecture I 6 Circadian Rhythm I 9 Neurological Basis of Sleep I 12 Age Relationships / 14 Sleep Needs I 15 Total Sleep Deprivation I 18 Deprivation of Specific Sleep Stages I 20 Dream Content I 21 22 Sleep Hygiene Effect of Exercise I 22 Effect oj Environment / 23 Effect of Food I 27 Sleep Medication I 29 34 Specific Sleep Disorders Nosological Considerations / 34 Primary Sleep Disorders I 35 Sleep Disorders Secondary to Medical Problems /53 Sleep Disorders Secondary to Psychiatric Problems I 55 Sleep Disorders Secondary to Behavior Problems / 58 Parasomnias I 63 69 Concluding Comments 70 Bibliography

I ~~~g;;r,~z:;:wl"f! Introduction

rent moods, electrodes and sensors are applied. At least eight electrodes are Between 12% and 15% of all people needed to determine sleep stages (Recht- living in industrialized countries have schaffen and Kales, 1968): two central- serious sleep problems. This 12% to 15% scalp electrodes (C3 and C4) to record an figure excludes the 20% to 25% of the electroencephalogram (EEG), two at people who complain of occasional bouts the outer canthi of the eyes to record eye of insomnia (Karacan et al, 1976). In a movements (the electro-oculogram, or careful survey of a Florida county, more EOG), two chin electrodes to record a i than 35% of the population reported mentalis electromyogram (EMG), and having trouble sleeping at least "some- two reference electrodes on the earlobes

times" and 13% said they had sleep prob- (Al and A2)· However, for a clinical as- lems either "often" or "all the time" sessment of sleep disorders, many more (Karacan et al, 1976). Similar estimates electrodes are usually required. For the I have been obtained ·in other studies evaluation of insomnia, for example. I from this country, from Britain, and one usually needs sensors that measure I· from Australia (Clift, 1975). Further- respiratory rate and breath-by-breath more, problems with insomnia are not air flow, an oxyrneter, EKG leads, and 1 the only sleep disturbances of concern. surface EMG leads over the right and I Excessive daytime sleepiness is a serious left anterior tibialis muscles. On the ! problem for well over 100,000 Americans other hand, a full clinical EEG is rarely (Guilleminault et al, 1974), and prob- obtained in sleep studies. lems during sleep, such as nightmares After electrode application, the sub- and enuresis, take a further toll. ject retires toa relatively quiet and com- The purpose of this review is to help fortable bedroom, which is separate clinicians effectively diagnose and treat from the equipment room but con- patients who complain of sleep prob- nected to it electronically and by an in- lems. Secondarily, it is hoped that crite- tercom system.A technician monitors ria for referring patients to sleep dis- the polysomnogram: the continuous orders centers will be established. To recording of EEG, muscle tension, eye achieve this goal, the review is divided movements, and other activities being into three parts: 1) an introduction to observed. Even though sleep labora- the basic facts of sleep; 2) a review ot the tories traditionally record at only 10 to factors that influence sleep; 3) a discus- 15 mm/sec, an entire night of polysom- sion of the diagnosis and treatment of nography generates from 1,000 to 1,500 the various sleep disorders. feet of paper. Although electronic techniques can The Sleep Laboratory be used to reduce the data, most sleep- Most knowledge concerning sleep in stage analyses are still done by visual humans has been gathered in sleep lab- inspection of the polysomnogram. The oratories, which usually function as fol- classification of sleep stages depends on lows. The subject, either a normal volun- multichannel patterns and relation- teer or a patient suffering from sleep ships rather than on individual EEG disorders, comes to the lab one or two frequencies, and electronic techniques hours before his usual bedtime. After are still less reliable and more expen- the subject fills out a questionnaire con- sive than is the trained eye for this com- cerning his daytime activity and his cur- plex analysis.

5 ~~~------~.------

"see" (that is. remember or recall) pic- Figun turcs presented to them in this stage (Rechtschaffen and Foulkes, 1965). - Sleepers also change mentally during Basic Facts on Sleep stage 1 as thoughts begin to drift. Think- ing is no longer reality-oriented. and Awak ~ short dreams often develop (Foulkes and Vogel, 1965). Nevertheless, many people Sleep Architecture subjectively feel that they are awake dur- ~ Less than 50 years ago, sleep was thought ing stage 1. Because stage 1 is similar to to be a relatively simple, uniform, and sleep in some respects (lack of reactivity homogenous state. The concept changed to certain stimuli) and similar to wake- in 1935, when Loomis et al described fulness in others (relatively desynchro- Drow; ¥=~ separate and distinct EEG stages within nized EEG), it probably should be con- sleep.The next significant advances sidered a transition phase rather than occurred in the mid-1950s. when Aser- regarded as full sleep (Johnson. 1973). ~tM insky and Kleitman (1953) reported that Stage 2 is marked by the appearance bursts of conjugate. rapid eye move- of sleep spindles (ie, bursts of 12 to 14 ments (REM) periodically appear dur- cycles per second [cps] activity lasting ing sleep and when Dement and Kleit- one half to two seconds) and by K-com- StagE man (1955) linked these periods of REM plexes (well-delineated, slow-negative to dreaming. So, sleep can be looked EEG deflections that are followed by a upon as a complex and multivariable positive component). Stage 2 is the first A/vI, state not unlike a building with various bona fide sleep stage, and mentation components and structural laws relating during this stage usually consists of ••• one part to another. Sleep architecture, short, mundane, and fragmented then, describes the components of sleep thoughts (Foulkes. 1962). (stages, cycles) and their interrelation- Delta sleep is defined by an EEG rec- Stage ships. ord showing at least 20% waves with a Sleep Stages: Basically, there are two frequency of one half to two cps or very different kinds of sleep: Non-Rapid slower and amplitudes greater than

Eye Movement (NREM, pronounced 75 microvolts (JtV) peak to peak (C3A2 or

Non-Rem) sleep, also called "orthodox" C4Aj). In the past, delta sleep has been sleep or "the Svstate," and Rapid Eye separated further into sleep stages 3 and Movement (REM) sleep. also called 4 depending on the number of delta "paradoxical" sleep or "the D-state" waves. but this differentiation has prov- Within NREM sleep. three stages are en unnecessarily detailed in most cases. Delta usually recognized: stage 1, stage 2, and REM sleep alternates with NREM delta sleep. sleep at about gO-minute intervals. The Stage 1 is a transition phase between EEG pattern during REM resembles full wakefulness and clear sleep. On the that of stage 1 sleep. except that saw- polysomnogram it is identified by a rel- tooth waves are often seen (Figure 1). atively low voltage, mixed-frequency The rapid eye movements (REM), which EEG with a prominence of activity in the give this stage its name, are only one three-to-seven eps range (Rechtschaffen of the outstanding characteristics. Mus- and Kales, 1968). In normal sleepers, de tonus is extremely low (especially stage 1 tends to be relatively short, rang- around the neck and chin), though many ing from one half to seven minutes. small twitches are seen during REM REM Reactivity to outside stimuli is dim in- sleep. Heart rate and respiratory rate ished.For exam pIe, college students are relatively high and very variable. with their eyes taped open no longer More than 80% of subjects awakened

i.~~ .~C:_ •• ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~:l .,

Figure 7. Human Sleep Stages

Awake - low voltage - random, fast

1 sec

Drowsy - 8 to 12 cps - alpha waves ,I ,I ,. J , Stage 1 - 3 to 7 cps - theta waves

...•. Stage 2 -12 to 14 cps - sleep spindles and K complexes

Delta Sleep - Yz to 2 cps - delta waves> 75 !IV I

tI>: i REM Sleep +low voltage - random, fast with sawtooth waves II· Sawtooth Waves Sawtooth Waves .--, r-----.

7 · " j"") 1

,~ during REM sleep are able to recall Following stage 2, the second REM pe- dreams (Snyder, 1971). riod of the night occurs about three Events during REM sleep are CUSW- hours after falling asleep. This second ,J' marily divided into two classes: tonic REM period lasts about ten minutes and events (such as low muscle tonus and is more intense (both physiologically and desynchronized EEG) that form the sta- psychologically) than the first. ble base of this stage, and phasic events Following the second REM period (such as eye movements and muscle and until awakening in the morning, twitches) that are superimposed in short stage 2 sleep and REM sleep alternate in bursts on this tonic background. about gO-minute cycles. Delta sleep is Sleep Cycles: Upon falling asleep, a rarely seen in these later sleep cycles. subject goes through a stage of relaxed REM periods become more intense wakefulness characterized by alpha (both physiologically and psychological- . waves. Later, the sleeper passes through Iy) and longer towards morning. The stage 1 into stage 2,Gradually descend- mean length of a REM period is' about ing deeper into sleep, most young adults 15 minutes, but REM periods frequently enter delta sleep within 30 to 45 minutes last up to one hour. after sleep onset. Depending on the sub- , Although the separation of sleep into jeers age, this sleep stage persists from a mutually exclusive, separate stages is few minutes up to an hour, then it yields convenient for data reduction, sleep again to stage 2 sleep.About 70 to 90 stages actually merge into one another. minutes after sleep onset, the sleeper be- Delta waves, for example, gradually begins the first REM period of the night; come more abundant and gain ampli- this REM period usually lasts about five tude following the onset of sleep.There minutes, It is by far the least-intense is no clear threshold where one sudden- REM period of the night, in terms of ly enters delta sleep (except by arbitrary both the physiological manifestations of definition). Similarly. the indices of REM REM sleep and the psychological inten- sleep are strongest in the middle of sity of dreams (Figure 2). the REM period, whereas the transition The second sleep cycle begins when point between stage 2 sleep and REM is stage 2 sleep recurs after the first REM often quite difficult to define. period. On some occasions, delta sleep Sleep is frequently interrupted by reappears, but there is usually less delta body movements and 8 to 15 short sleep in the second cycle than in the first. awakenings, even in those who sleep

Figure 2. Typical Sleep Pattern of a Young Human Adult Awake ••

Stage 1 and REM

Stage 2

Stage 1 sleep and REM sleep (black) are graphed on the same level because Delta tneir EEG patterns are very similar. Hours 2 3 4 5 6 7 8

8 e., _--~.... ~. --we-

well. There seems to be an optimal num- tuations occur around this mean.Tem- ber of body movements (around one peratures are low during sleep, rise in every 15 to 20 minutes), and Othmer the morning, and reach a high sometime (1965) feels that too few body move- during a person's period of peak effi- ments might be as detrimental to good ciency. Many other bodily functions, sleep as too many. especially endocrine secretions and Depth of Sleep: Of the NREM stages, metabolism, follow similar circadian delta sleep is deepest and stage 1is light- rhythms (Weitzman et al, 1968; Weitz- est (if it is sleep at all).However, REM . man et al, 1976). sleep cannot be that easily classified on a When human beings are removed scale of sleep depth. By measuring how from all indications of time, ie, being much noise it takes to wake a person placed in an underground cave with from REM sleep, it appears that REM artificial light, plenty of food, but no in humans is about as deep as stage 2 clock, they still maintain a circadian sleep (Rechtschaffen, Hauri, and Zeitlin, rhythm. However, under such constant 1966). In the cat, however, REM is the conditions the circadian rhythm rarely "deepest" sleep of the night, ie, it takes lasts exactly 24 hours. Usually humans more noise to awaken a cat from REM fall into rhythms between 24 and 28 sleep than from delta sleep. Similarly, hours, though sleep-wake cycles of up to while such things as heart rate variabil- 50 hours have been observed. ity and neural discharges in some parts In order to entrain (synchronize) the of the brain suggest that REM sleep circadian rhythm to the exact 24 hours might be as energy consuming as alert imposed on us by the sun,"Zeitgebers" wakefulness, a lack of reactivity in skin (ie, indicators of time) are necessary. resistance, an exceedingly low muscle Such Zeitgebers may be clocks, time of tonus, and many other measures suggest wakening by alarm, time of regular that REM sleep might be extremely meals or work times, or position of the deep sleep. sun (Aschoff et al, 1975). Some data even Because REM sleep cannot be classi- suggest that naturally occurring, ex- fied as either light or deep, some re- tremely low-frequency electromagnetic searchers now believe that there are fields (at 10 cps) might be Zeitgebers three separate stages of existence: wake- . (Wever, 1974). fulness, sleep, and the REM state. The Excessively Long Circadian Rhythms: REM state behaviorally is similar to normal sleep (one lies in bed with eyes closed Case History: Miss R., a 26-year-old, single = journalist, sough; help at the clinic for sleep- '" and diminished responsivity to the en- onset insomnia. She explained thai she would vironment); however, it is much more often lie in bed four to six hours hefin» fa llin g similar to wakefulness in other respects asleep each night, and then would have extreme (desynchronized EEG, arousal in many difficulties getting up the next morning (she physiological systems). In short, REM had recently been fired from her job because she appears to be a state as different from ouerslept so rrgularly), Houievrr; ('xCI'PIfor a NREM sleep as it is from wakefulness. certain dejensiueness cunrrrning her ilwbility to get up on time, Miss R. seemed to be in reason- Circadian Rhythm abl» good mental health. "Circa" is Latin for "about; and "dian" Miss R. was asked to keep a sleep log at home for two weeks. This log revealed that now, while means "day." Circadian rhythms are she was unemployed, her sleep-onset insomnia those fluctuations that take about 24 had totally disappeared. Instead, each night hours to complete. Miss R. now went to bedfour to six hours later The Healthy Rhythm: Normal body than she had the previous day, causing her to temperature is said to be about 36.8 C. :~Iay up all night and sleep during the day on but regular and consistent 24-hour flue- certain occasions. AI this point, Miss R. kept

9 herself occupied by doing research for a book she and forth every few days, and a consist- planned to write. She was counseled to continue ent circadian rhythm is never established. this activity. irrespective of the time of day or Manipulating the sleep-wake rhythm night, to go to bed only when she felt sleepy. and is largely a behavioral issue. One can to sleep as long as she wished. A subsequent sleep force only wakefulness; one cannot logfilled out for more than a month was astound- force sleep. If habitual oversleeping is ing: Miss R. felt healthy and alert, usually working 20 to 22 consecutive hours, then relax- a problem, going to bed earlier when ing for one or two hours before sleeping uninter- not tired will only result in sleep-onset ruptedly for 10 to 12 hours (Figure 3). Realizing difficulties. It will probably not result how well she felt under this new regimen, Miss in a phase shift. However, forcing per- R. decided to become a freelance writer and to sistent arousal at a specific time each continue "free running" on a 36-hour to 38- morning will finally result in a phase hour day. She has been on this schedule for over shift, by causing tiredness earlier in the six months now, without ill effects. evening. Similarly, arousal too early in Although most of us show circadian the morning cannot be corrected by try- rhythms somewhat longer than a nor- ing to force oneself to sleep longer. Only mal 24-hour day in a timeless environ- by forcing oneself to remain awake later ment, we are apparently able to syn- in the evening can one gradually ac- chronize with the sun quite easily. How- complish a phase shift, over a period of ever, the possibility exists that some weeks, and sleep longer in the morning. people (such as Miss R.) might be unable Lack of Zeitgeber: to compress their circadian rhythms into the usual 24-hour day. Case History: Mr. M., a 25-year-old me- chanic, had been hit lly a train and was in a coma Manipulation of Circadian Rhythms: for three months. He then made a dramatic ~. The position of the peaks and troughs of recovery, and seemed to be normal except for a the circadian rhythm can apparently slight aphasia. His lawyers, however, advised him ~I be adjusted within the 24-hour day. This not to work until the damage suit was settled. is often done passively by flying across Mr. M. had always been a "loner," and he a number of time zones. After a few spent the next two years in a small apartment days of discomfort ('jet lag") when our doing very little else except reading. Whenever biological circadian rhythm is out of he needed food, he would get itfrom a nearby 24- synchrony with the local time, the body hour grocery store. Thus, while there were the adjusts to its new environment but is usual Zeitgebers in his environment, Mr. M. then out of synchrony with home (Me- ignored them. As a result, he soon found that he Farland,1975). started to sleep poorly and always felt tired. After two years on a "timeless" behavior The position of peaks and troughs schedule, Mr. M. was referred to the clinic as an within the 24-hour clock time can ap- insomniac. Twenty-Jour-hour recordings at the parently be manipulated by an act of clinic revealed that he took frequent short naps will, as is done in shift work. A worker throughout the day and night. To our surprise, who consistently stays up all night and body temperature showed no discernible cir- goes to bed at 8 AM learns [0 sleep until, cadian rhythm (Figure 4), but only wide and say, 4 PM. Under these conditions, body random fluctuations. temperature will be low at noon and NIr. M. was urged to reestablish a regular day- high at midnight, and other circadian night rhythm. When that proved impossible be- events will also adapt to this new time cause he always [e lt tired, he hired afriend to live with him to keep him awake and active during pattern. However, adjustment to the the day, while gradually increasing his time newly imposed sleep-wake rhythm may awake between 7 AM and 10 PM. Regular take a week or more.The problem with mealtimes and a slow increase in exercise were shift work is that on the days off, the also prescribed. Over a period of six very difficult worker tries to become a "day person" months. Mr. M. reestablished a normal circa- again. Thus, the Zeitgebers switch back dian cycle, and felt more fit during the day.After

10 "

Figure 3. Excessively Long Sleep-Wake Rhythm

Example of a normal sleep-wake rhythm

Awake Working

Awake Relaxing I I I Sleep I Thursday

Free-running, sleep-wake rhythm of Miss R

Awake Working

Awake Relaxing I

Sleep

Saturday Sunday

Figure 4. "Regular" and "Irregular" Insomnia

Regular sleep-onset insomniac

Sleep in hours

99- Oral 98- temp. 97-

Insomniac (Mr M) lacking a consistent sleep-wake' rhythm

Sleep 1V2 4 2 1V2 5 3V2 1V2 4 in hours

99- Oral 98- temp. I 97-

Saturday Sunday Monday Tuesday Wednesday

II the settlement of the lawsuit IY2 ,vears later, Mr. wake rhythm and to resynchronize the M. bought a smallfarm in New England, which various circadian cycles (Lund, 1974). he t\ now managing quite I!jjl!cth1ely.The in- .Because [he major Zeitgebers in our cul- somnia has totall» disappeared. ture are social (work. meals, evening Case History: When Mrs.B. came to the sleep activities), scheduling problems such as clinic. she was almost totally crippll'd by her the ones discussed in theabove two case chronic insomnia. She felt too weak to walk histories are seen mainly in people who further than to the bathroom. usually uienl lo bed are not under pressure to follow them around 8 PM, and then stayed in bed until 5 rigidly, such as schizoid, withdrawn per- PM the following evening. What little sleep she sons and those who are unemployed or got came in short episodes of one half to two self-employed. hours irregularly spaced throughout the day and Obviously. establishing circadian night. She arose-o-n-ly- to keep he> husband rhythms on the basis of oral tempera- company during the dinner hours, and managed everything else (including a weekly women's ture is scientifically weak. Measuring grouP) while lying in bed. temperature rectally, or assessing me- History revealed that Mrs. B. had always been tabolism, would be more convincing. a "fickle" sleeper but her main trouble started However, for clinical purposes, oral about 15years ago when she had gone through a temperature is much easier to obtain traumatic experience involving, within a period and often convincing enough. Graphing of six months, the death of her son, the loss of her this temperature over a period of weeks husband's business, and her removal from her gives the patient a powerful incentive to towns "high society. Sleeping poorly at night reestablish a regular sleep-wake rhythm. because of the trauma, she first started to escape The shape of the 24-hour oscillations her trouble by taking afternoon naps. Sleeping worse and worse at night, and feeling weaker seems partially related to personality and weaker, she extended her time in bed until variables. "Early birds" reach the peak of she stayed there for 22 hours per day. biological function and performance Careful medical evaluation showed no spe- earlier than "night owls;' who show very cific findings, except for muscular atrophy and little increase in metabolism and per- weakness apparently secondary to prolonged bed formance for hours after awakening rest. The depression of 15 years ago had long from sleep, but then reach their peak since lifted. H oureuer; both body temperature later during the day (Blake, 1965). and endocrine function had lost the usual 24- hour cycling, both showing erratic, wild fluctua- Neurological Basis of Sleep tions. Mrs.B. was encouraged to increase the time out of bed very gradually and to become more The search for neurological regulators involved, once more, in her husband's business of sleep has a long and distinguished and town affairs. Over a period of three years of history. As early as 1890, Mauthner spec- constant counseling and support, she learned to ulated that the area around the nucleus stay up from about 8 AM to 10 PM and has of the third cranial nerve might be re- regained her old position in civic affairs. Night laced to sleep induction. His conclusion sleep consolidated. H otueuer, M TS. M. never was based both on postmortem exami- became an excellent sleeper and, after very poor nation of patients who had suffered nights, she still takes a 1Y2-hour afternoon nap from a form of encephalitis with sleepi- once or twice per week. ness and on theoretical speculations. Both Mr. M. and Mrs. B. had lost a Reasoning from his now famous en- clear 24-hour pattern in their lives, each cephale isole and cerveau isole exper- for a different reason. By the time the iments, Bremer (1935) concluded that original reasons (hospitalization for Mr. the junction between the diencephalon M., depression for Mrs.B.) had disap- and the brain stem must be crucial to peared, circadian cycling had given way sleep and wakefulness. Like Mauthner, to erratic fluctuations. It was then very Bremer also felt that sleep occurred "by difficult to reestablish a healthy sleep- default," that is, after afferent impulses

12 to the cortex fell to a level too low to main- debatable. It may be located either in the tain wakefulness. A modified version of serotonergic nuclei of the pontine raphe this view gained widespread credence system or in the medial forebrain area. when Moruzzi and Magoun (1949) pro- For the REM state to occur, the active posed the existence of the reticular acti- involvement of the nucleus ceruleus vating system for maintaining arousal. (NC) and the gigantocellular tegmental Powerful evidence contradicting the field (GTF), both located in the higher above idea that sleep was purely the pons, seems necessary (Figure 5). absence of wakefulness was presented Based on a series of brilliant experi- by Hess (1943, 1954). Hess felt that his ments in cats, Hobson and co-workers experiments indicated the existence of (eg, Hobson et al, 1975) propose that an active sleep-inducing center, because cells in the gigantocellular tegmental he could actually "force" animals into field (GTF) produce electrical dis- sleep by his stimulations. Lesion experi- charges spontaneously, unless such dis- ments by Nauta (1946) supported this charges are inhibited by impulses from view. The arguments between "passive" the nucleus ceruleus (NC),As will be sleep (based on a lack of stimulation) seen below, this inhibitory relationship and "active" sleep (induced by activity in between NC and GTF seems crucial for a sleep-promoting system) then raged REM sleep. When GTF cells discharge, for over two decades until it became they set up spikes that travel into many clear, in the late 1960s, that both active different neurological systems. They are and passive theories were partially right then often called pontine-geniculate- (as will be discussed below). occipital spikes (PGO spikes), because A separate breakthrough in under- they were first found to travel from the standing the neurophysiology of sleep pons to the lateral geniculate nuclei, and came around 1960, when Jouvet and his from there to the occipital cortex. co-workers (1959, 1962, 1965) located a The PGO spikes may be the origin REM-inducing system in the higher and the prime movers for most phasic pons. and some tonic phenomena during REM As understood today (Hobson, 1974), sleep. Such peo spikes are found enter- our wake-sleep-REM life depends on ing the oculomotor nuclei, possibly giv- three neurological systems. These sys- ing rise to the rapid eye movements of tems should not be thought of as specific REM sleep. They appear in pathways to nuclei,or centers, but rather as networks the middle-ear muscles, travel to the of mutually interacting and inhibitory midbrain reticular formation (where areas or circuits. they may be associated with the tonic Wakefulness seems to be maintained EEG desynchronization typical of REM largely through activity in the ascend- sleep), and they seem related to many ing reticular-activating system. As long aspects of the high variability found in as' activity in the system is high, there the autonomic nervous system during cannot be any sleep. Low activity levels REM sleep - for example, instability of in this system apparently lead to relaxed the heart rate. They may even be in- wakefulness (often indicated by alpha volved in the psychological aspects of waves) and possibly to stage 1 (the transi- REM dreaming. Discharges coming from tion phase between wakefulness and the lateral geniculate nuclei may be in- sleep). terpreted by the occipital cortex as visual Sleep (stage 2 and delta) requires the pictures coming from the retina. active involvement of a hypnagogic or The biochemical regulation of human sleep system in addition to low activity in sleep probably involves numerous dif- the reticular-activating system. The lo- ferent ncurotransmitter substances. cation of this hypnagogic system is still Some tantalizing bits of evidence have

13 already emerged. In the cat, the neuro- arousal in the reticular activating system. transmitter involved in the hypnagogic Chronic insomnia, however, especially circuitry emanating from the pontine when it has existed since early infancy, raphe is serotonin. Although things might not always be related to excessive seem more complicated in humans (Wy- arousal. Rather, one might suspect a att, 1972), Hartmann et al (1974) have direct weakness in the sleep-inducing cir- shown that as little as one gram of cuitry or in its biochemical components, tryptophan, the precursor of serotonin, as will be discussed later. can reduce sleep latency. In addition to the work on neuro- The neurotransmitter substance transmitters, Pappenheimer (1976) and that inhibits gigantocellular tegmental Monnier et al (1972) seem to have isolat- field activity seems to be aminergic ed two different complex chemicals in (norepinephrine or serotonin, or both). the CNS and blood that, when injected This would explain why medications into recipients, apparently cause sleep. that increase aminergic activity at the How these chemicals relate to the neuro- relevant pontine sites usually decrease logical mechanisms discussed above is REM sleep. When such drugs are with- unknown at present. drawn abruptly, the neurochemical bal- Age Relationships ance is usually disturbed in the opposite direction, and there is more REM sleep Age seems to be the single most power- · . per night for up to four weeks after ful determinant of a person's sleep pat- ~. withdrawal from certain drugs. This tern (Williams et aI, 1974). Total time REM sleep is often more intensive, both spent in bed drops from about 17 hours physiologically and psychologically (Os- per 24 hours near birth to about 81'2 wald, 1969). The frightening dreams oc- hours at age 12 and to about 71'2 hours casionally experienced after abru pt with- between the ages of 25 and 45. From drawal from drugs that influence the then on, total time spent in bed rises aminergic system have been explained again, until it reaches a mean of about as being results of this REM rebound. 81'2 hours in old age. Physicians might do well to caution The average time asleep per 24-hour their patients about the possibility of period also drops from about 16 hours such frightening dreams when with- per day for neonates to about 8 hours at drawing them abruptly from drugs that the age ofl2 before leveling out at about increase aminergic activity (eg, hyp- 7 hours between the ages of 25 and 45. notics, stimulants, antihistamines, anti- From then on, total sleep time continues depressants). to decline gradually, until it reaches Most of the neurophysiological and about 6 Y2 hours in old age (Figure 6). neurochemical details remain to be Comparing total sleep time with total worked out, but nobody questions the time in bed, it is clear that sleep efficiency existence of three neurological systems (time asleepltime in bed) remains rela- governing wakefulness, sleep, and REM. tively stable up to about45 years of age, From the interaction between the sleep but then starts to drop.The older people and the wake systems, it follows that in- get, the longer they stay in bed but the somnia can occur either through exces- less sleep they obtain per 24-hour period. sive activity in the wake system (ie, the With advancing age, sleep also becomes reticular-activating system) or through more fragmented - more awakenings some inadequacy in the sleep circuitry. during the night and more difficulties Although scientific data are lacking, it returning to sleep once awakened. appears that the occasional insomnia The proportion of time spent in vari- most of us experience around stressful ous sleep stages also changes quite dra- events relates almost exclusively to matically over a lifetime. Delta sleep

14 item. cially ancy, .ssive ~ct a g cir- ents,

~uro- land .olat- lis in ected leep. ~uro- ve IS

Figure 5. Schematic Diagram of the REM System iwer- , pat- time From the Nucleus Ceruleus (NC) to where they are apparently interpreted as coming lours the Gigantocellular Tegmental Field (GTF) from the retina (seen as pictures in our dreams). t8Y:! a.Discharges from the NC usually inhibit dis- c. Because the pathway GTF-LGN-OC was the charges from the GTFduring waking and NREM sleep. first one discovered, the discharges are called iours b. The neurotransmitler involved in this inhibition pontine-geniculate-occipital spikes, or PGO spikes. 'rom appears to be aminergic. Other discharges from GTF nses c. This inhibition breaks down just before and during The discharges from the GTF enter a multitude of REM, liberating the GTF cells to discharge spikes. ibout other neuronal structures, apparently causing many of the typical phenomena related to REM sleep. hour From GTF via LGN to OC; The PGO Spike Only two are outlined here: lours a. Discharges from the gigantocellular tegmenlal a. Discharges enter the oculomotor nuclei (OMN) field (GTF) enter the lateral geniculate nucleus (LGN), and cause eye movements. irs at a halfway station on the visual pathway from the b. Discharges enter the midbrain reticular formation ibout retina to the occipital cortex (OC). (MRF) and cause general desynchronization of the d 45. b. Via the LGN, these discharges then enter the OC, cortex typical for REM sleep. mues iches 'e 6). hovers around 15% to 20% up to the age seems to be a slight percentage increase total of about 20 but then gradually decreases, during young adulthood and a slight iency until there is practically no delta sleep percentage decrease, at least in males, rela- left by the age of 60. Conversely, stage 1 during old age. The meaning of these age, "sleep" increases from about 5% in child- percentage changes in REM sleep is un- .ople hood to about 15% of sleep in old age. known, but they have given rise to many t the Both of these changes imply that sleep speculations concerning our "need to riod. becomes lighter and less efficient as one dream" and the function of REM sleep. »nes grows older. lIngs REM sleep changes more dramatical- Sleep Needs ilties ly than any other stage during child- Individual Requirements: The preced- hood, dropping from about 50% at birth ing section established the average levels vari- to about 20% by puberty, from that time of sleep requirements over a lifetime. dra- on, REM sleep remains reasonably con- However, not all people need the same .leep stant throughout life, although there amount of sleep, and the acceptable

15 Figure 6, Mean Development of Human Sleep Over a Lifetime i, '" \, 16-"

14- Awake

12-

g- 2- QJ u; 2::- co u ro ,§

This graph shows changes (with age) in total amounts daily REM sleep falls steadily during life, the percent- of daily sleep, daily REM sleep, and in percentage of age rises slightly in adolescence and early adult- REM sleep. Note sharp diminution of REM sleep in the hood. This rise does not reflect an increase in amount: early years. REM sleep falls from 8 hours at birth to less it is because REM sleep does not diminish as quickly than 1 hour in old age. The amount of Non-REM sleep as does total sleep. throughout life remains more constant. falling from 8 hours at birth to 5 hours in old age. In contrast to the (Reprtnted by permission from Dr. H. Roffwarg. Montefiore Medi- steep decline of REM sleep, the quantity of Non-REM cal Center, NY, revised since publication in Science 152:604- sleep is undiminished for many years. Although total 619,1966.)

16 range, seems to be surprisingly wide. variety of psychological and social prob- Some healthy adults sleep three hours lems. In contrast, Hartmann's short or less per night without apparent ill sleepers, like Mrs. C., appeared to be effects, and a few rare ones have been "successful, outgoing, and healthy." recorded who habitually sleep one hour How much sleep does a "normal" indi- or less (jones and Oswald, 1968; Meddis vidual need?Anyone can establish his or et al, 1973). At the other extreme, some her own required quota by experiment- healthy adults sleep lO to 12 hours per ing with different sleep lengths, provid- night and feel sleep-deprived and tired ing each specific length is tried for at when they can obtain "only" seven hours' least a few weeks before trying another sleep per night. This wide range in the length. If any specific length does not sleep needs of apparently normal adults permit adequate sleep, one will become makes it necessary to define many sleep excessively tired over the course of a problems individually and subjectively. week or two, ready to drop off to sleep A person has insomnia if his inability to when not stimulated. With too much sleep interferes chronically with efficient day- sleep, one will find oneselflying in bed, time function, regardless of how many hours developing symptoms of insomnia. This, he sleeps each night. Thus, some insom- however, assumes that the person is free niacs might sleep seven hours, whereas from significant sleep pathology. As we some normal sleepers might sleep only will discuss later, some patients with ex- three. This definition is important, be- cessive daytime sleepiness remain sleepy cause some people complain bitterly no matter how many hours they rest. about their insomnia when, in fact, they Extreme differences in the need for just have a low need for sleep. sleep characterizes children as well as Case History: Mrs. C. was 71years old when adults. Under no circumstances should referred to the sleep clinic because she had never the average sleep times discussed in the slept more than three hours per night throughout section on age relationships be taken as her adult life. She felt well during the day, al- absolute norms for individual children. though she speculated she might feel even better Many two-year-old children need less if she could obtain a "normal" amount of sleep. She was physically well, still cross-country skiing sleep than ten-year-old children. The every day of the winter, and hiking frequently yardstick for adequate sleep in child- during the summer. A n extensive medical work- hood is that the child remains alert up, as well as neuropsychological testing, was and energetic over weeks and months. totally unremarkable except for some slight de- Should the child become excessively terioration commensurate with her age. tired or doze off at school or play in re- I Laboratory recordings confirmed an average sponse to sleep curtailment, he or she of three hours' sleep per night. Surprisingly, apparently needs more sleep. However, Mrs. C. spent more than a full hour in delta should even drastically increased sleep, ie, her sleep was much more sound than amounts of sleep time not bring alert- would have been expected for her age, Interview- ness and well-being after a few nights, ing Mrs. C. in detail indicated that she felt un- I concerned about her "insomnia" but had agreed the diseases of excessive daytime sleepi- to a sleep evaluation at the urging of her hus- ness (eg, sleep apnea, narcolepsy, to be band and her physician, both of whom had discussed later) need to be investigated thought that she was an insomniac. She was re- in such a child, as they would in the assured that she was a "healthy hyposomniac" adult. rather than an insomniac and seemed pleased Manipulating Sleep Needs: Are we with this diagnosis. chronically sleep deprived? Webb and It is not known why humans vary so Agnew (1975) found that students now widely in their sleep needs. Hartmann seem to sleep about 112 hours less per (1973) claims that long sleepers appear 24-hour period than they did in 1910, to be worriers, nonconformists, with a but that many would be willing to sleep

17 longer than their customary time if giv- sleep are unknown. We do not know, for en the opportunity. Webb and Agnew example, the effect of sleep curtailment believe that sleep is under increasing on psychological development, resist- prcs:;ure from modern living, jammed ance to stress, or the cardiac system. tightly between rigid work schedules, in- Total Sleep Deprivation creased opportunities to socialize in the evening, television, and electric lights- Effects: Can some people exist without and they suggest that most of us might any sleep? Many who claim to do this ac- do well to sleep longer. tually take short naps when studied in While this view might apply to the the laboratory. However, in certain very busy, successful fellow who can never rare forms of organic brain disease (eg, find enough time to "catch up on his Morvan's chorea), sleep seems totally sleep," sleep clinicians usually warn abolished (Fischer-Perroudon et al, about the contrary problem - staying in 1974), and the patients survive, at least bed longer than is necessary, which can for a few months. lead to behavioral insomnia (see page Keeping healthy, young volunteers 58). "Sunday night insomnia" is a com- awake for an entire night makes them mon complaint. After having slept ex- very sleepy but has remarkably little cessively long on Saturday and Sunday effect on their performance during the mornings, it apparently becomes diffi- next day (johnson, 1969; Wilkinson, cult for many people to fall asleep at 1969; Morgan, 1974). Initially, this was their regular bedtime on Sunday night. a very surprising finding because, after Perhaps keeping a little "pressure" in the a sleepless night, mood deteriorated !!.. sleep system would facilitate ease of fall- markedly and volunteers presumed that ing asleep. Assuming that most humans they were performing poorly. Except have an inherent circadian rhythm some- for extremely monotonous tasks, one what longer than 24 hours, this would can apparently "rally" one's reserves and !". make good sense: getting up somewhat function about as well after a sleepless before the system is ready to get up' night as after a good night's rest. might help to squeeze our inherently After two or three nights of total sleep longer circadian cycle into the 24 hours deprivation, however, small "micro- required by the clock. sleeps" (beginning with a few seconds' Can we decrease our need to sleep? duration) intrude into wakefulness. The answer seems to be a qualified "yes" These microsleeps cause occasional, (Mullaney et al, 1976). Three couples short inattention to the task at hand, and who habitually slept eight hours per performance then deteriorates even un- night and one who usually slept 67"2 der high motivation. If the task allows hours were asked to reduce their total for such short periods of inattention (eg, sleep time by 30 minutes every two self-paced tasks), the performance dec- weeks until they reached six hours, and rement after extended sleep depriva- by 30 minutes every three weeks from tion is smaller than if the task requires then on. One couple terminated the ex- constant vigilance. Microsleeps become periment at 5 \12 hours, two stopped at 5 longer and more frequent as total sleep hours, and one at 4\12 hours. They were deprivation continues beyond three then allowed to sleep ad lib again. At a days. After about ten days, microsleeps one-year follow-up, six of the volunteers are so intertwined with wakefulness that still slept 1 to 2 \12 hours less than they it becomes almost impossible to deter- had before the beginning of the experi- mine whether a person is actually awake ment. Although there were no visible or asleep, even if he performs functions signs of distress at the one-year follow- usually associated with wakefulness (eg, up, the long-range effects of curtailing walking). Apart from extreme sleepiness

18 and these microsleeps, however, there wakefulness shows an increase in delta is remarkably little abnormal pathology sleep and, on the first recovery night, of- associated with extended stints of sleep- ten shows a decrease in REM sleep. Re- lessness.This was demonstrated by Randy covery sleep need not be equal in length Gardner, who showed no demonstrable to total sleep loss, and it usually cannot abnormality after staying awake for 264 be taken all at once. Even after ten days hours (Dement, 1972). of total sleep deprivation, a person rare- The fact that laboratory volunteers ly sleeps more than 14 to 20 hours at any can function quite well after one to three one stretch, and two-or three nights of nights of sleep deprivation contrasts extended recovery sleep usually will re- with the feelings of patients with insom- turn a healthy volunteer to normal nia, who believe that they cannot func- (Kales et al, 1970;Williams et aI, 1964). tion at all without their full quota of In summary, an occasional night of sleep. Psychological components seem very poor sleep (or no sleep at all) seems to play an important role in this differ- of little medical concern in healthy in- ence.The volunteer enters a sleep- dividuals. However, short-term sleep deprivation experiment at a reasonably losses may be of concern in certain types convenient time prepared for the sleep- of epilepsy (lack of sleep appears to lessness. Besides the excitement, there lower seizure threshold), in medically is the challenge to perform well after marginal patients, in jobs where an occa- sleepless nights and to see how long one sionallapse of attention might be critical can remain awake. In real life, however, (eg, air-traffic controllers), and possibly sleepless nights usually come at the most in jobs that depend on one's mood (eg, inopportune times, for example, before receptionists). an important test or before a board Therapeutic Deprivation: Surprising- meeting. The night is spent in frustra- ly, although total sleep deprivation dis- tion, anxiety, and anger at oneself. This turbs the mood of normal individuals, in psychological handling of a sleepless patients who suffer from endogenous night seems to contribute markedly to depression, a totally sleepless night im- the general feeling of malaise the next posed by the therapist seems to cause day. Similarly, if sleeplessness is caused psychiatric improvements (Schulte, 1959; by a medical problem (eg, a developing Van den Burg and Van den Hoofdakker, cold, an allergy), the concurrent effects 1975). Unfortunately, however, such im- from the illness contribute to the gener- provements last only a few days. al feeling of malaise from sleeplessness. Why should a night of total sleep Obviously, chronic sleeplessness is not deprivation work as a treatment for en- the same as spending a few nights in a dogenous depression?The reason is sleep-deprivation experiment. Chroni- currently unknown. Perhaps, like other cally poor sleepers facing anew a totally "shocking" procedures, loss of sleep sleepless night might already have de- temporarily startles the patients, rallies pleted their reserves and might be less their reserves, and snaps them out of able to "rally" a second wind. Neverthe- their depression. On the other hand, less, it seems important to let insomniac Pflug (1972) points out that both sero- patients know that they can function tonin and norepinephrine imbalances more or less adequately after one poor are related to endogenous depression as night, even though they feel moody well as to sleep. Possibly, a dramatic and not up to par. This knowledge often alteration of sleep might cause a tempo- helps them sleep better by eliminating rary change towards the better in the some of the panic poor sleepers experi- disturbed neurotransmitter balances of ence when facing a sleepless night. endogenous depression, whereas the Recovery: Recovery sleep after long same alteration might wreak havoc with

19 the optimal tuning of the balances in whenever the polygraph indicated the onset of normals. REM sleep On the fourth evening, he did not appear at the lab. Because he had reliably come Deprivation of Specific Sleep Stages to the lab exactly on time, an extensive search for REM: When volunteers are REM-de- him was undertaken. Two hours later, he was prived by awakening them each time finally found attending a nearby burlesque they start a REM period, "REM pres- show. According to him, he had never before even considered enjoying himself in that way. He sure" seems to build up. Upon falling claimed that after the three nights of REM dep- asleep again, they usually wait less than rivation, he suddenly felt different and didn't the usual 1Y2 hours before they start care what was going to happen or what he was REM sleep again (eg, only 45 minutes af- supposed to do. ter the first REM awakening and 30 min- In the early 1960s, it was thought that utes after the second). They also make up total deprivation of REM sleep for a few the lost REM time as soon as they are per- nights might lead to psychosis. This is mitted to do so (Figure 7). In addition, not true. However, as was true of Mr. D., REM sleep becomes more intensive (eg, people who are REM-deprived seem to more phasic eye movements) when it is become more agitated, more impulsive, under pressure, and on awakening, cer- and less superego-controlled, whereas tain behavioral changes occur. people deprived of delta sleep do not Case History: Mr. D., a college student, had . show these signs (Agnew et al, 1967). · volunteered for a study of REM deprivation in Observing less superego control in our lab. He was an extremely conscientious per- normal individuals deprived of REM son making A + grades (which he earned mainly sleep, and considering that most anti- by studying exactly what was required for each course, handing all assignments in on time, etc.). depressant medications severely restrict He seemed to have total control of himself and to REM sleep, Vogel et al (1976) suggested give himself no time at all for pleasurable, non- that the build-up of pressure for REM academic pursuits. Mr. D. was REM-deprived sleep might be the critical factor in the for three consecutive nights by being awakened treatment of endogenous depression.

Figure 7. REM Deprivation and Rebound

REM suppressant I .. I I --- medication administered ------Placebo administered ---

40- REM suppression

30- Q. Mean REM level CD for this person CD u; :2 w a: .S 20- 1 C CD Q. U) Q. CD Pre-drug Steady state Steady state ~ 10- steady reached during drug reached again o state administration under placebo S c t CD (J ( Co ~ O-~------~~ t Time (successive nights)

20 Vogel et al deprived 17 hospitalized, en- sleep seems related to psychological re- dogenous depressives of REM sleep for covery. three weeks by awakening them at the on- Dream Content set of each REM period. An equal number of similar patients were awakened Dream analysis is beyond the scope of the same number of times from NREM this review. Nevertheless, a few basic sleep.The REM-deprived group im- points concerning dream content might proved significantly more (were less de- be made, because patients often insist on pressed) than did the NREM-deprived telling bothersome dreams to their phy- group. On the other hand, patients with SICIans. reactive depression did not respond very No matter how strange, a dream is well to REM deprivation. Vogel et al also created by the dreamer's own fantasy present evidence that endogenous de- and thoughts. Thus, a dreamer might pressive patients who do not respond to do well to take responsibility for his REM deprivation fail to respond to imi- dreams, try to understand them, not fear pramine as well but can be treated with them or laugh them off as "nonsense" electroconvulsive therapy (ECT). Follow- We seem to dream mainly about issues up data suggest that Vogel's "REM-dep- that concern us at the present time, such rivation cure" is at least as efficient as are as fears, wishes, plans, hopes, worries. the usual drug treatments. Thus, dreaming that a friend is dead Not only do Vogel's studies suggest does not necessarily signify a latent new ways of treating depression, they death wish but might well mean that one also raise theoretical questions concern- is concerned for the friend's health. ing the mechanisms involved both in de- Dreams often relate to "unsettled pression and in REM sleep. In addition, business" stirred up during the day. For Vogel's studies demonstrate that "nor- example, a glimpse of a girl may remind mal" sleep might not always be the best someone, for a split second, of a former kind for everyone. This is important to girl friend. If engaged in conversation, remember in the evaluation of drug ef- the person may disregard this split- fects on sleep. It is commonly assumed second thought. That night, however, that drugs that do not disturb the bal- issues relating to the girl friend may ance between sleep stages are preferable prompt dreaming, even though the wak- . , to those that do, but Vogel's work chal- ing stimulus for the dream cannot be re- lenges that a priori assumption. called. If the issues related to the girl Delta: Volunteers deprived of delta friend are still a matter of concern or sleep react differently than do volun- are otherwise not properly "settled," teers deprived of REM sleep. Rather it is even more likely that such a dream than becoming agitated and more im- will take place. pulsive, as occurs after REM deprivation, The language of dreams is often sym- delta-deprived persons become physi- bolic and distorted. An apple might cally uncomfortable, withdrawn, and stand for a forbidden object; a wild tiger manifest concern over vague physical for rage. Disentangling the "real" mean- complaints and changes in bodily feeling of the dream, therefore, needs de- ings (Agnew et al, 1967). Also, muscles tailed and intimate knowledge both of become more sensitive to pressure, and the dreamer and of dream mechanisms volunteers can withstand less musculo- in general. Nevertheless, it is often bene- skeletal pain after delta deprivation than ficial to ask patients who insist on telling they could on the preceding evening their dreams how they interpret them. (Moldofsky et al, 1976). This suggests In this way important concerns that that delta sleep might be related to the patient is unable to voice directly rrrusc ulosb elet.al recovery, whereas REM often surface and can be dealt with.

21 A second problem relates to individual differences. One man's relaxation .is another man's stress. Some need absolute quiet to fall asleep, whereas others do best when listening to music or when the TV is blaring. Robinson (1969) found that good sleepers slept even better after two hours of studying just before bedtime. On the other hand, the sleep of usually poor sleepers deteriorated even further after such studying. Individual differences appear to be especially prominent if the various environmental factors take on personal meanings. Many "fresh-air fiends" sleep well with windows wide open on a 40-degree spring night, but let the house get that cold in the middle of the winter, and they awaken from sleep and complain bitterly about freezing. A third problem relates: to the self- fulfilling prophecies that are so common Sleep Hygiene in ~leep work. A patient who expects to sleep poorly in a certain environment often worries a}Oout sleep and does poorly as a result f it. Expecting a good

Folklore abounds with do's and don'ts night's sleep in ~o certain environment concerning ideal sleep arrangements. reduces worry afd this, in itself, con- Should we sleep on a hard mattress, in a tributes to good seep. cold room, on an empty stomach, alone, Effect of Exercis or with a partner? Before looking into these issues, some words of caution are Everyone knows how soundly people necessary. sleep after exercisy. Or dowe?The ques- Most of us sleep poorly in an unfamiliar tion is more complex than originally environment. This is known to sleep re- thought, but an a~swer is important for searchers as the "first-night effect." It theoretical as well as practical reasons. means that sleep is usually disturbed in If sleep "restores~ur batteries;' as many good sleepers on the first night in the believe, we woul seem to need more, laboratory. On the other hand, poor or more efficien , sleep after exercise sleepers often sleep excellently on the than after relaxation. Here are the facts first night in the lab, and some reasons as we know them today: for this paradoxical finding will be dis- A chronic, con~tant amount of exer- cussed on pages 58-63. What seems more cise does seem te benefit sleep, since surprising than the initially disturbed athletes seem to hr,ve more delta (deep) sleep in a new situation is the speed with sleep than nonat~letes (Baekeland and which good sleepers can adapt to unfa- Lasky, 1966; Zlot et ai, 1973). Increas- miliar environments. In any case, before ing exercise loads for one day in non- deciding whether a certain arrangement athletes does not sfem to increase delta is disturbing or helpful to sleep, the sleep on the foHewing night (Hauri, sleeper needs to be thoroughly adapted 1968; Zir et al, 19111),but denying ath- to it. letes their customary exercise decreases

22 ,.

the amount of their delta sleep (Baeke- sleep) is needed after each of the two land and Lasky, 1966). conditions.Second, one might speculate Exercise seems most effective in influ- that slow EEG waves are merely an epi- encing sleep if it is done in the afternoon phenomenon, not the "essence" of the or early evening. Exercise in the morn- deep, sound sleep we have-come to asso- ing has little effect, possibly because the ciate with the delta stage. Slow EEG waves ensuing wakefulness is sufficient for and some presumably more basic proc- physiological recovery (Horne and Por- esses of delta sleep might usually show a ter, 1976).Excessive exercise immedi- high correlation, yet the association ately before sleep seems to cause arousal between the two events might be lost and stress, which might interfere with under atypical conditions such as sen- possible beneficial effects of sleep (Hob- sory deprivation or bed rest (Zubek et al, son, 1968). 1963). In any case, there are conditions, The amount of peripheral, physiolog- such as certain organic brain syndromes, ical arousal following exercise influences where slow EEG waves do not imply neither sleep onset nor sleep stages in delta sleep. . healthy young adults. Studying the ef- The issue does not seem resolvable at fects of evening activities on subsequent this time. Nevertheless, in terms of sleep sleep, Hauri (1968, 1969, 1970) had hygiene, the following recommenda- students exercise intensively for six tions can be supported by current re- hours on one evening and relax for six search findings: hours on another evening. Surprisingly, • A steady amount of exercise is probably there was no difference in the ease with beneficial for sound sleep, but one cannot which subjects fell asleep after the two force sleep on any given night by exercising evening activities, nor in the develop- excessivel y durin g the precedin g day. Rather, such excessive exercise may result ment of sleep stages through the first in increased aches and pains that might three hours of the night. Sleep came detract from sleep on the following night. just as easily after exercise as it did after •If one exercises in order to sleep well, relaxation, even though heart rate, res- such exercise should probably come in the piratory rate, and rectal tern perature afternoon or early evening, not in the morn- were still significantly higher at sleep on- ing or just before sleep. set after exercise than after relaxation. Effect of Environment No condition seems as diametrically opposed to exercise as prolonged bed Noise: How much noise can we tolerate rest, yet two studies (Ryback et al, 1971a, before our sleep becomes disturbed? To 1971b) have shown that bed rest of five answer this question, one needs to qualify weeks' duration increased the amount of it in at least six different ways. delta sleep per night. This startling find- First, the awakening threshold de- ingcontradicts all commonsense notions pends on the stage of sleep. Stage 1 is about a direct relationship between the lightest stage; that is, it takes the delta sleep and physiological recovery least amount of noise to cause full arous- after exercise. al, whereas stages 2 and REM are deeper, Keeping in mind that these studies and delta sleep is deepest (Rechtschaffen were performed on healthy volunteers, et al, 1966). However, human REM sleep there seem to be two possible explana- can be very deep if the external noise is tions. First, there are only two ways of somehow incorporated into the dream dramatically changing muscle mass, content. Bradley and Meddis (1974) either by hypertrophy (following exer- measured an average REM arousal cise) or by disuse atrophy. Perhaps a threshold of 60 decibels on the A level similar intensity of physiological recov- (dBA) in their sleepers when the noise ery and repair (as evidenced by delta they fed into the sleeper's room was not

23 incorporated into the dream but an discusses such a mechanism in detail, average threshold of 70 dBA when it was but there has been little factual research incorporated. They gave the following carried out on this point. example of a dream that incorporated Fifth, the age of the sleeper seems the noise and which had, therefore, a crucial. Although hearing usually be- high arousal threshold: comes less acute with age, the sleep of We had to ask for more stationery, and you older patients seems to be more sensitive have tofill in a form to ask for it and that noise to noise than does that of younger peo- started and mvfattier did not know how to stop ple. Lukas and Kryter (1970) present it, and he picked up a file and flicked through evidence that older sleepers awaken it, lookingfor thisform and when he came to the much more frequently to sonic booms form that would stop the noise ... but there wasn't than do younger sleepers. Similarly, a form in the file. So that noise wouldn't stop. Roth et al (1972) found that the same Second, the auditory awakening noise level that awakened 70 year olds threshold depends on how much sleep caused only a temporary shift toward a person has already accumulated during stage 1 sleep in 25 year olds. the night. The same stage, later in the Finally, noise sensitivity is different in night, is lighter than it was earlier in the men than in women. Dobbs (1972) estab- •.. night (Rechtschaffen et ai, 1966). lished that women awaken more easily to Third, individual differences in sensi- airplane noise than do men. tivity to noise seem.very important. Some Keeping these qualifications in mind, people awaken more easily than others, we can now examine the effect of noise, even from the same sleep stage. In the particularly truck and aircraft noise, on study by Rechtschaffen et aI (1966) men- sleep.Most studies have used young tioned above, the average awakening adults, who may be subjected to simu- threshold throughout the night was 15 lated aircraft noise or truck noise dur- dBA for one sleeper and more than ing some laboratory nights, to no noise 100 dBA for another. during other nights. Unfortunately, While examining individual differ- this procedure resembles having a coun- ences in arousal thresholds, Zimmerman try man suddenly try to sleep close to a (1970) found that light sleepers showed busy airport. Rarely is enough time given faster heart and respiratory rates than for adequate adaptation. Nevertheless, did deep sleepers, higher body tempera- Chiles and West (1972) found no evi- ture during sleep, more spontaneous dence that one sonic boom per hour awakenings, and more body movements during sleep produced any measurable while asleep.These differences between consequences on a complex task per- light and deep sleepers are similar to formed the next day. On the other hand, those observed by Monroe (1969) be- Levere et al (1972), using nine simu- tween poor and good sleepers. Zimmer- lated jet aircraft overflights per night man, therefore, speculated that his light (20-second duration, maximum inten- sleepers might be the population "at sity 80 dBA), found that daytime perform- risk" for developing insomnia later on. ance on a complex monitoring task de- Fourth, the specific meaning of the clined after the flyover nights, in spite noise seems important. The mother who of the fact that normally the noise had sleeps through thunderstorms but awak- not awakened the subjects. In the morn- ens when the baby fusses in the next ing, many of them were unable to recall room is well known. Apparently, a "night a single flyover, Nevertheless, the sleep light of consciousness" is still burning EEG of these subjects characteristically during sleep, discriminating among had shown a marked desynchronization incoming auditory stimuli and deciding in response to the noise, indicating a on appropriate responses. Morgan (1970) movement towards lighter sleep stages.

24 This desynchronization lasted, on the airport still showed less delta sleep, more average, for five minutes after the noise wakefulness, and more stage 1 sleep had disappeared. In a comparison be- than did residents living in quiet neigh- tween sonic booms and truck noise, Berry borhoods. In fact, residents close to the and Thiessen (1970) found that the airport recorded 45 minutes less "use- longer-lasting truck noises at 70 dBA ful" sleep (stage 2, delta, and REM) per produced more awakenings than did night than did residents living in quiet the shorter and more impulsive sonic neighborhoods, and the disruptions booms at 120 dBA. and awakenings in their sleep could be Herbert and Wilkinson (1973) stimu- directly traced to aircraft flyovers. lated ten subjects during one of five Would it help to attenuate the noise nights with clicks of variable loudness by the use of earplugs or to mask it by a (65,75,80,90 dBA) administered every constant noise such as an air conditioner? 20 seconds. These clicks caused a signifi- Otto (1972) found that fiberglass ear- cant increase in stage 1 and wakefulness plugs (with about 20 dB attenuation) and a trend towards less REM and less helped solidify sleep in the face of nor- delta sleep. However, the effect of such mal street-traffic noises. As for screen- a sleep disturbance on awakening per- ing, Scott (1972) found that a constant formance (measured with vigilance white noise of 93 dBA decreased REM tasks and addition tests) was relatively and increased stages 1 and 2 sleep, al- small and confined to the early waking though it did not change the number hours. of awakenings or the amount of delta Studying the problem of human adap- sleep when compared with quiet nights. tation to noise for extended periods, Because Scott studied only one night, Cantrell (1974) had 20 young males con- it is unclear whether one might grad" fined to a dormitory for 55 consecutive ually habituate to the white noise over days. During 30 of these days, high- successive nights. That such habituation pitched tones at 80, 85, and 90 dBA were can occur during sleep has been estab- emitted through omnipresent loud- lished (johnson et al, 1975; McDonald speakers at 22-second intervals, 24 hours and Carpenter, 1975), although adapta- per day. Although the volunteers com- tion is much slower during sleep than it plained that the 85 and 90 dBA tones is during wakefulness. interfered with their ability to fall asleep, In summary, loud noise does disturb this complaint could not be substantiated sleep, even in people who do not awaken in the laboratory. However, delta sleep from it, do not remember the noise in I ~ _1 did decrease during this relatively pro- the morning, or supposedly are habitu- i! longed and stressful time. Townsend et al ated to it. Noise usually shifts sleep away Ii (1973), in a similar experiment, found from delta and away from REM, and it ~J similar results. increases the number of awakenings I, i The most direct answer to the ques- and the amount of stage 1 sleep. Muf- tion of whether chronic noise interferes fling excessive noise that can reach the with normal sleep comes from Globus bedroom seems to be a good idea for et al (1973), who studied the sleep of most people. However, whether a spe- long-term Los Angeles residents in their cific noise is disruptive to the sleep of a own homes. Half of the volunteers in specific person depends largely on indi- this study lived adjacent to Los Angeles vidual differences, such as age, sex, the International Airport; the other half meaning of the noise, and the physical resided in quieter Los Angeles neighbor- properties (pitch, duration, and ampli- hoods. Even though each of the subjects tude) of the noise. had resided in the same neighborhood Physical Surroundings: Throughout for at Ic:llst six years, those: close: to the history, bed preferences have ranged

25 -. r from the hard floor to the extremely soft beds of the 18th century in which one slept in a near-sitting position. Similarly, many Europeans today sleep with the head part of the mattress elevated by a special triangular supplement, whereas Americans sleep on flat mattresses. There is no evidence that such differences affect sleep once one is accustomed to them. Are the shape and the quality of the mattress important to sleep? Kleitman (1963, p 309) pointed out that the curvature of the spine does not seem to be in- Table I Ten Rules to Better Sleep Hygiene fluenced by how straight one lies in bed: "Aside from the fact that the spine has a 1. Sleep as much as needed to feel refreshed couple of natural curves, there is no and healthy during the following day, but not evidence that the spine curvature is more. Curtailing time in bed a bit seems to solidify sleep; excessively long times in bed changed in the cat, which sleeps curled seem related to fragmented and shallow up, in the Japanese, who sleep on the sleep. ground, or in sailors, who often sleep in 2. A regular arousal time in the morning seems bona fide hammocks with a consider- to strengthen circadian cycling and to able sag in the middle." Well-constructed finally lead to regular times of sleep onset. mattresses maybe important for people 3. A steady daily amount of exercise probably with orthopedic problems; they seem deepens sleep over the long run, but occasional one-shot exercise does not directly unnecessary for most normal sleepers. influence sleep during the following night. It has been established that excessively 4. Occasional loud noises (eg, aircraft flyovers) hard surfaces (ie, wooden floors) can disturb sleep even in people who do not cause more body movements, more awaken because of the noises and cannot awakenings, and more stage 1 sleep than remember them in the morning. Sound do softer surfaces (Kinkel and Maxion, attenuating the bedroom might be advisable for people who have to sleep close to ex- 1970). Beyond this, however, sleep seems cessive noise. intrinsically very little affected by the 5. Although an excessively warm room disturbs characteristics of the sleeping surface, sleep, there is no evidence that an exces- except for matters of personal prefer- sively cold room solidifies sleep, as has ences and style. Even such exotic situa- been claimed. tions as air-fluidized bead beds seem to 6. Hunger may disturb sleep. A light bedtime have few lasting effects on sleep in nor- snack (expecially warm milk or similar drink) seems to help many individuals sleep. mal subjects (Dement et al, 1971;Shur- ley, 1971; Rosekind et ai, 1976). 7. An occasional sleeping pill may be of some benefit. but the chronic use of hypnotics is If the type of mattress used does not ineffective at most and detrimental in some influence sleep, as long as it is reason- insomniacs. ably comfortable, why should patients 8. Caffeine in the evening disturbs sleep, even sleep better on air-fluidized beds (Kline [. in persons who do not feel it does. et al, 1974) or waterbeds?As shall be ~, 9. Alcohol helps tense people to fall asleep fast. discussed later (p 60), some patients are but the ensuing sleep is then fragmented. conditioned against relaxing in their own 10. Rather than trying harder and harder to fall beds. In these patients, the more the new asleep during a poor night, switching on the sleep situation differs from the old one, light and doing something else may help the individual who feels angry, frustrated, or tense the less conditioning can carry over from J. about being unable to sleep. the old to the new situation. Add to this 26 "A' . '<5 l,~ ------~I !i a healthy dose of placebo when one and animals as well as humans become ! changes to a more expensive mattress, poikilothermic (Baker and Hayward, II and it may well be that sleep improves 1967; Shapiro et ai, 1974).Sleeping in quite dramatically in specific cases. extreme temperatures might, therefore, In a study conducted by Monroe (1969), have less effect in the early part of the I couples who were good sleepers and night when REM periods are short, but ,( habitually slept in the same bed were it might possibly become disastrous il studied in the lab. They slept together on later on when REM periods become II some nights, apart on others. Monroe longer. - found more delta sleep when the couple Weather: Popular thought has long i~ slept apart, and somewhat less REM sleep. maintained that specific weather condi- .~ Sleeping together may be good for mari- tions could influence sleep.Sleep is often [ tal bliss, but sleeping apart does seem to said to be especially poor when a front is , deepen sleep by preventing disturbances approaching or when particularly dry t when the partner changes position. winds prevail, such as the Santa Ana f Temperature: Ideal sleep tempera- in Southern California or the fohn in t tures differ for different species: the rat Switzerland and Austria. Ii apparently sleeps best at 30 to 32 C(86 Some scientific evidence now sup- I~ to 90 F) (Valatx et aI, 1973), whereas cats ports the popular association between ~ appear to sleep longest at around 22 C weather and sleep. Raboutet et al (1959) l~ (72 F) (Parmeggiani and Rabini, 1970). found a statistically significant relation- i Unfortunately, an ideal temperature for ship between barometric pressure and " humans has not been determined to sleep onset as assessed in routine EEGs I~' date. A study of sleep differences in done on aviation personnel: the higher I j. psychiatric patients sleeping at 18 C, the barometric pressure was, the fewer I i 23C, and 29C(64F, 74F, 84F) found no were the people who fell asleep during a I[ consistent trends toward better sleep in clinical EEG. Webb and Ades (1964), fol- I a particular temperature (Presley et al, lowhingup on RIahb~)Uhtet'sdlead,founld tlhat I 1973). The old belief that sleeping is bot extreme y Ig an extreme y ow !!'! best when the bedroom is cool, therefore, barometric pressures were associated still awaits scientific -validation.It has with increasing sleepiness during clinical been determined, however, that room EEGs. temperature will influence dream content. As the room temperature was de- Effect of Food creased from 22 C (72 F) to 17 C (63 F) Caloric Intake: Losing or gaining weight and later to 12C (54F), dreams became often has surprisingly profound influ- more emotional and more unpleasant ences on sleep. In general, weight gain is (Ziegler, 1973). associated with long, uninterrupted Sleeping in a hot room (temperature sleep, whereas weight loss is associated much above 24C [75F]) seems to be bad with short and fragmented sleep. This for humans: they wake up more, move has been demonstrated most convinc- more in their sleep, and both REM and ingly by Jacobs and McGinty (1971) in delta sleep are decreased (Schmidt- rats: the less food their rats received, Kessen and Kendel, 1973; Otto et ai, the less they slept, until after 6 to 11 days 1971). This is probably why Kleitrnan, of food deprivation, virtually all sleep as early as 1937, found more body move- disappeared. The survival value of a ments and more interruptions in the food-sleep association is obvious: Starv- sleep of his Chicago subjects during the ing animals ought to be awake and look- summer than during the winter. ing for food rather than sleeping; sati- It appears that during REM sleep, aied animals are probably safest if they thermoregulation is seriously impaired, can hide and sleep for long periods. Pa-

27 renthetically, one wonders whether in tion between food and sleep might be our weight-conscious society a fair amount secondary to thyroid function. Hyper- of poor sleep might be directly related to thyroidism is associated with weight loss semistarvation diets. and short but intensive sleep (much Evidence for an association between delta), whereas hypothyroidism results food intake and sleep in humans comes in weight gain and long sleep with a rela- mainly from psychiatric clinics.Irrespec- tive lack of delta (Kales and Kales, 1974). tive of their diagnosis, psychiatric out- The association between food intake patients who have recently lost weight and sleep is not invariably present; some usually complain about poor sleep and studies have been unable to confirm more awakenings, especially in the second the above findings. Nevertheless, chronic half of the night. Patients who recently insomniacs and those awakening too have increased their weight usually sleep early in the morning might do well to try long and soundly (Crisp and Stone hill, a midnight snack in an effort to solidify 1973). In addition, insomnia is a com, their sleep .. mon symptom in anorexia nervosa.(Lacey Tryptophan and Sleep: Postprandial et al, 1975); when such patients are fed, drowsiness after a good meal is a com- their sleep improves. Similarly, in demon phenomenon.Similarly.warm milk, pression, it appears that most patients Ovaltine, and Horlicks (a Scotch malt both lose weight and become insomniacs, drink) all improve sleep significantly although some depressed patients in- when taken before bedtime (Kleitman crease their weight and sleep for exces- et ai, 1937; Brezinova and Oswald, 1972). sively long times (Michaelis and Hof- It appears that these phenomena might mann, 1973). not only be related to caloric intake (as Why should food intake be related to discussed above), but specifically could sleep? Fara et al (1969) introduced milk affect the ingestion of L-tryptophan, or fat directly into the duodenum of cats the serotonin precursor, that exists in and observed improved sleep. They many foods. As discussed before, sero- ,I· postulated that sleep was induced either tonin appears to be related to sleep mecha- by the direct, central action of a released nisms, and Hartmann et al (1974) have gastrointestinal hormone or by indirect demonstrated that as little as 1 gm of neurogenic mechanisms triggered tryptophan can clearly improve sleep in through the stimulation of duodenal some patients. receptors. On the other hand, it seems The production of serotonin depends quite possible that some of the associa- on the availability of tryptophan which is

Figure 8. Stage 2 Sleep Patterns Before and After Withdrawal From Hypnotics (Mr. D.) Note the mix of spindling and alpha waves before withdrawal, and the relative absence of alpha waves and the reappearance of K complexes after withdrawal.

Before drug withdrawal Alpha Waves Alpha Waves

Alpha Waves Spindle Spindle Spindle

Atter drug withdrawal

K Complex carried into the brain on a transport and social ties to famil)1 and friends were re- protein. Unfortunately, the transport established in joint therapy. A low-grade chronic protein has a lower affinity for trypto- depression was treated with 100 mg doxepin phan than for four major competing h.s. Six months later Mr. D. started to function amino acids. Thus, increased amounts of adequately once more. He still sleeps poorly but does get up more refreshed. Laboratory studies tryptophan will enter the brain only if nOlll show about five hours of normal sleep but blood concentrations of tryptophan are still a high number of auiahenmgs (Figure 8). increased relative to the concentration of the competing amino acids (Wurtman Mr. D. obviously suffered iatrogenic and Fernstrom, 1974; Kolata, 1976). insomnia. His case is not rare. Most sleep clinicians know patients who received a Sleep Medication hypnotic for a temporary disturbance Hypnotics are not entirely harmless. (such as surgery) but from whom the Carelessly prescribed over months and drug was never withdrawn. Over the years, they apparently can destroy sleep, years, dosages gradually crept upwards as the following case history illustrates. until natural sleep became impossible Case History: Mr. D., 58 years old, operated a (Kales et al, 1974). small business. He complained of inability to Franklin (1969) demonstrated that an sleep even though he took at least 40 mg of diaz- entire psychiatric admission ward can be epam, 90 mg of jlurazepam, and 400 mg of managed quite well with very little use of secobarbital each night. When this mixture did hypnotics, once the staff has changed its not help, he often used up to afifth of whiskey to attitude towards sleeping pills. Unfortu- induce sleep. He was referred to the sleep clinic nately, in the United States, the average after hisfourth "blackout episode." According to complaint of insomnia is dealt with in less Mr. D., blackouts occurred on the nights when than three minutes by prescribing hyp- he took "whole handfuls" of sleeping pills in a desperate attempt to find some rest. Four times notics (Dement, personal communica- after such episodes, his wife found him lying on tion). This is worrisome, because many the jloor unconscious and was unable to arouse patients who initially complain about him; she then rushed him to the hospital. On one sleeplessness do not suffer from insom- occasion, he remained in a coma for jive days nia when evaluated in the lab. In these before recovering. instances, extensive probing may be nec- A sleep evaluation revealed that Mr. D.'s essary to uncover their basic psychiatric habitual nightl)l~dose of hypnotics produced or medical reasons for seeking help. absolutely no normal sleep. Although the EEG Others, although true insomniacs, suffer did slow somewhat for about two hours, and a from problems not treatable with hyp- few eye movements around 3 AM suggested a weak attempt at a REM period, nothing in notics (as shall be discussed below). Mr. D.:, record looked like normal sleep. Nev- Although the overprescription of hyp- ertheless, Mr. D. claimed that his lab sleep had notics may be an important cause of been "much better" than a normal night at home. chronic insomnia, the informed use of Mr. D. related that he had been a good sleeper hypnotics is essential in the management until two weeks after Pearl Harbor, when he of sleeplessness. Some rules for such a had become plant manager. Being ill-prepared use of hypnotics will therefore be devel- for thejob, he then started to worry and to sleep oped here without, however, discussing poorly, and a barbiturate was prescribed. Over the details of each prescription hypnotic the next 30 years, his sleep gradually deteriorated, and he required more and more hypnotics and over-the-counter sleeping pill. Care- and alcohol to reach a state of oblivion. ful reviews comparing specific hypnotics Mr. D. was admitted to the hospital and grad- have recently been published (johns, ually withdrawn from all hypnotics. He was 1975; Greenblatt and Miller, 1974). In taught electromyogram (EMG) biofeedback controlled studies, hypnotics are more and relaxation exercises. A lcoholu:s A nony- alike than different. Most show basically mou.! became involved because of his drinking, the same course (Figure 9): an initially

29 ·REM Jt n r- I II nn n r1 III II I

II II III II r • I In '-1 I III III I

0- I II n n rr III

III I II Ii II In n r I In II II I r

II II I ""'1 r I r I

Hours 2 3 4 5 6 7 8

Figure 9. Development of Sleep in a 17-Year-Old Insomniac on 100 mg Secobarbital

a. Typical sleep cycle of a young adult (taken from Figure 2). b. Sleep of the 17-year-old insomniac on placebo -long sleep latency, fragmented sleep. c. Sleep on the first drug night - note solid sleep and long delay of REM sleep. d. Sleep on fifth drug night - insomnia has returned: long sleep latency. fragmented sleep. eo SI99p on the lirst drug-withdrawal night: sleep latency of 3'1.1 hours, fragmented sleep.

30 satisfactor y j reactio n that diminishes hypnotics are most effective on the first over time. Also, most patients show a ..re- night of use, but some, with long half- bound insomnia" when withdrawn from lives of their active components, are more their hypnotic. Such a pattern is graphed effective on the second drug night. in Figure 9: A young adult with sleep- 3) What level of sleep will beproduced with onset insomnia was recorded in the clinic, chronic use? In Mr.D.'s case (above), the then 100 mg secobarbital was prescribed. chronic administration of hypnotic drugs While on this medication, the patient eventually seems to have produced less seemed to sleep well the first night, but sleep than would have been obtained after as few as five nights, the insomnia without drugs. Other drugs apparently " returned. When the drug was discontin- level out at the patients' average sleep I, ', ued after one week of chronic use, the time, letting the insomniacs sleep about patient initially slept even more poorly as poorly with the chronic use of a hyp- than beforethe medication was given. notic as without it. No drugs have yet Clinical Efficacy: More important been shown conclusively to improve than theseemingly objective efficacy of sleep permanently over baseline condi- a hypnotic is the subjective feeling of tions. having slept well, the ability to function 4) What is the immediate effect of with- well during subsequent wakefulness, drawal? When a hypnotic that has been and the lack of any medical and psycho- taken chronically is precipitously with- logical damage that might be caused by drawn, total sleep times of less than one the hypnotic. A drug that could consist- half hour per night are common. It is ently produce a good feeling during then often imagined that such a poor wakefulness, while maintaining sound sleep would be the patient's lot forever psychological as well as medical health should he withdraw from the medica- on a long-term basis, would be the ideal tion. This erroneous assumption pro- drug. Thus, the main questions when' longs unnecessary dependency on hyp- evaluating hypnotics are "Does the pa- notics. However, care is indicated when tient feel better and more alert during withdrawing medication. A precipitous the day following the use of hypnotics? withdrawal from certain hypnotics can Does he do so without incurring a debt cause serious medical sequelae, such as in terms of side effects and/or medical the seizures induced by sudden barbitu- and psychological sequelae?" rate withdrawal. . The following five efficacy questions 5) How long do the withdrawal effects are also crucial when evaluating a hyp- last? Most studies document withdrawal notic (see Figure 10): effects only for three or four days, but 1) How effective is the drug on the first patients often complain of sleeping .... night? What is the minimum adequate poorly after drug withdrawal for up to dose .for securing a night of adequate a month. In the laboratory, Haider and sleep, and how close is this dose to a toxic Oswald (1970) found that overdosages one? of certain hypnotics resulted in abnor- 2) How long is the drug effective? Some mal sleep patterns for up to two months drugs (eg, secobarbital, 100 mg) lose after complete withdrawal. their sleep-inducing potency within one Nobody has yet documented the long- week of chronic administration (Kales range effects of hypnotic drugs when et al, 1976). Others (eg, fturazepam) lose used for months, years, or decades, even efficacy more gradually, although they though this may be a serious health ques- are still better than is the placebo after tion. Obstacles to studying this problem one month of chronic administration are enormous. It is difficult to assess pre- (Kales et aI, 1975). In addition, the shape drug levels of sleep and psychological of the curve seems important. Many function in those who have taken hyp-

31 notics over a long period of time, and to While these procedures make intuitive do the needed prospective study would sense, no acceptable research has been strain finances and resources consider- done to date on this topic of "drug holi- ably. Nevertheless, the costs of not know- days,' a serious and glaring blind spot in ing the answer may, in the long run, far our knowledge. outweigh the costs of finding out, both A careful evaluation of the effects for individuals and for society as a whole. hypnotics have on the different sleep Before using a hypnotic drug, each stages is not included in this discussion. patient should be familiar with the basic According to Johnson (1973), there is curve as outlined in Figure 10. A patient little difference in the recovery value of might be advised, for example, to take different sleep stages, at least in terms the hypnotic each night for the next of subsequent performance. Similarly, two weeks because of specific job pres- some humans can apparently function sure, providing he can then take some quite adequately without any REM sleep days off from work to accommodate (Wyatt et al, 1973), and the well-docu- the "insomnia-rebound" which inevitably mented lack of delta sleep following the will follow withdrawal. Similarly, an administration of benzodiazepines does insomniac might decide to take the medi- not seem to have much effect on wakeful- cation during weekdays but not on week- ness, either. The only documented con- ends, or on alternate days, or only during cern in the area of hypnotics and sleep two out of three weeks. stages is the REM rebound. Withdrawal

Figure 10. The Five Efficacy Questions for Hypnotics'

, j 1 3 5 How long do ! How effective What level of sleep will , ' is the drug on be produced with chronic use? the withdrawal !':' the first night? (In the case graphed here, effects last? the level lies lower than the nondrug level) 2 4 How long is the What is the immediate drug effective? effect of withdrawal?

9- A

:c Ol 'c Qj 3- Mean sleep before 0. any medication 0. Q) Q) 2- en 0 sn 1 - 5 ..• 0 I a - Start hypnotic

lime (successive nights) 'See page31

'. from most hypnotics usually results in antidepressants seem especially indicated an increase in and intensification of if the patient is depressed as well as in- REM sleep. This REM rebound may somniac, and several authorities have DUSt::: increased and unpleasant dream- suggested giving all of the antidepressant ing, even nightmares, and patients should medication in one dose at bedtime (Kara- be warned about this possibility when can et al, 1975). However, Fletnenbaum withdrawing from hypnotics. (1976) warns that combining all antidepressant medication into one single Agents Used to Induce Sleep dose at bedtime apparently did lead to Although iTIanydifferent hypnotics are some terrifying nightmares in his pa- currently marketed, the general trend tients. Such nightmares are rare, how- recently has been towards using the ever, if very low doses of sedating anti- benzodiazepines whenever possible. depressants (eg, 25 mg amitriptyline) This trend has a rational basis, although are administered for the sole purpose it should be clearly understood that the of solidifying sleep or abolishing alpha benzodiazepines are not yet the ideal intrusion into sleep, as will be later dis- drugs, and we do not know the effects of cussed (see page 51). these drugs when used for months or Over-the-counter medications advertised years. Also, benzodiazepines do not in- as soporiphics usually contain scopola- duce sleep in all insomniacs. mine and/or methapyrilene, which make Some of the rational reasons for using a patient groggy. If the patient's main benzodiazepines are: problems are worries about not sleep- • Long-range effectiveness (up to one ing, feeling this grogginess apparently month) as a hypnotic has been demon- relaxes him because he thinks that "the strated for some benzodiazepines, but sleeping pill has already started to work" not for the other sleeping pills. This thought alone will make it easier for • Fatal overdoses are much less com- many to fall asleep naturally. However, mon with benzodiazepines than they are over-the-counter drugs do not induce with such drugs as barbiturates, glutethi- sleep directly and are useless in more- mide, ethchlorvynol, chloral hydrate. serious insomnias (Kales et al, 1971). • Benzodiazepines as hypnotics are Aspirin is widely used to induce sleep. easier to withdraw than are other hypnot- In some cases, simple reduction of pain ics, apparently in part because the half- may be the explanation. However, aspirin life of some active metabolites in this also increases the proportion of free type of drug extends beyond 24 hours. tryptophan in the blood. Inasmuch as Thus, sleep is still somewhat aided by free tryptophan can bind to the transport the benzodiazepines on the first with- protein and cross the blood-brain bar- drawal night. rier, and tryptophan bound to serum • Barbiturates, glutethimide, and albumin cannot, it seems quite possible possibly ethchlorvynol potentiate certain that aspirin might directly improve sleep liver enzymes-and therefore influence by increasing brain serotonin levels. No the activity levels of drugs given concom- adequate study of this hypothesis has itantly. Antihistamines potentiate anti- been done to date. cholinergic drugs. Chloral hydrate influ- Although most hypnotics are unsatis- ences protein binding. Benzodiazepines, factory in the long run, some break- on the other hand, do not seem to inter- throughs may well be expected in this act significantly with other medications. area within the next few years. Hope- Sedating tricyclic antidepressants have fully, we shall learn more specifically been advocated by certain clinicians be- how to alter certain neurotransmitter cause they appear to maintain effective- balances at the appropriate sites, rather ness longer than do the hypnotics. Such than attempting to induce sleep by an

33 overall depression of CNS activity. Hope- Specific fully, we shall also learn better how to get the desired drug directly to specific Sleep Disorders areas of the brain rather than having to saturate the entire body. Four Rules for the Use of Hypnotics Nosological Considerations • If a person is under short-term stress Only during the past ten years have mainly manifested by sleeplessness, a the problems of sleep pathology been hypnotic drug to help him sleep for a few attacked by the technology of sleep re- nights seems indicated. However, care search (eg, Kales and Kales, 1970; Kara- should be taken to withdraw this hypnotic can et al, 1971; Oswald, 1969). Numer- either when the stress has terminated, or ous ways of classifying sleep problems no more than two to four weeks after the have been suggested since then, but initial prescription, depending on the none have yet gained universal accept- type of hypnotic prescribed. Should the ance. Currently a subcommittee of the stress manifest itself in daytime anxiety, Association of Sleep Disorders Centers a general tranquilizer seems more ap- is working on the classification of sleep propriate than the often-prescribed disorders, but their deliberations are not hypnotic. yet concluded. The classification system • Many insomniacs have a tendency used in this report follows the most com- to panic when they feel a series of poor monly used nosological categories, but ~:!r nights descending. A prophecy of a poor adds a category of "disorders secondary ;;r. I"~ night then becomes self-fulfilling as the to behavioral problems; which has not - I, ' panic increases agitation. Such patients been previously proposed.

.! .. should be given a few effective sleeping Until recently, sleep disorders were pills but admonished not to take more classified into the insomnias (not enough than one or two per week. The mere pos- sleep), the hypersomnias (too-much sleep), session of a hypnotic has a very relaxing and the dyssomnias (miscellaneous dis- and reassuring effect in such patients, turbances occurring during sleep, such and the sleeping pills will often not be as enuresis and sleepwalking). Logical needed. as such a classification might seem, the • Each patient using hypnotics should more one works with it, the less satis- be familiar with the basic relationships factory it becomes. Issues that should as outlined in Figure 10 so that he can be grouped together (eg, central sleep use his hypnotics wisely. apneas, resulting primarily in insomnia, •Many patients use low doses of hyp- and upper airway apneas, resulting pri- notics for years without undue side ef- marily in hypersomnia) are split apart fects and without needing to increase the in such a system. In addition, many sleep dosage. The placebo effect of such sleep- disorders belong in more than one cate- ing medication is apparently enough to gory. Narcolepsy, for example, is usually induce sound sleep. There seems to be classified as a form of hypersomnia be- little reason to categorically withdraw cause excessive and inappropriate sleep such patients from all hypnotic use.How- attacks during the day are its main symp- ever, patients who continually need to toms. However, most narcoleptics also increase their medication while remain- sleep very poorly at night and might just ing dissatisfied with their sleep probably as well be classified as insomniacs. Worst should be withdrawn from all hypnotics of all, excessive daytime sleepiness, the and treated behaviorally, by other hallmark of hypersomnia, is also an ex- pharmacological means (such as antide- pected outcome of chronic insomnia. pressants) or by psychotherapy. Thus, hypersomnia and insomnia have

34 .--i¥ic -

the same chief complaint of excessive than seven, has insomnia, but a person day lime sleepiness, and thi~ has led to who needs and obtains five hours of unnecessary semantic arguments. sleep each night is not an insomniac. The nosology used here first separates Excessive Daytime Sleepiness (EDS) has problems into primary and secondary sleep replaced the term "hypersomnia" in disorders. Primary sleep disorders (as most circumstances.This is so primarily defined by Williams et al, 1974) include because many people formerly classified the disorders in which disturbances or as hypersomniacs do not actually sleep abnormalities of sleep are the principal much longer than normals when mea- or only symptoms of the problem (eg, sured over 24 hours. Rather, their main sleep apnea, narcolepsy, primary insom- problem is that they chronically feel nia).Secondary sleep disorders are those sleepy, no matter how many hours they in which the sleep problem is only part allot to their sleep. of a symptom complex belonging to a As will become clear, knowledge more widely ranging clinical problem. needed to treat sleep disorders involves Among the secondary sleep disorders, general medicine, endocrinology, neuro- the current review classifies them along chemistry, neurology, otolaryngology, traditional lines: sleep disorders second- and cardiology, as well as psychiatry, ary to psychiatric problems, secondary psychology, and other behavioral sciences. to medical problems, or secondary to be- Historically, these disciplines have shown havioral problems. Clearly, these three little interest in the clinical problems of categories are not mutually exclusive. sleep. A new type of specialist has there- Poor sleep secondary to alcoholism, for fore appeared, during the past few years, example, could be classified into anyone who is primarily concerned with the of the three categories (it has arbitrarily diagnosis and treatment of sleep-related been put into psychiatric problems in disorders. this review).Also, for purposes of simpli- It seems too early to tell whether or fication, sleep-exacerbated disorders not we are now witnessing the emer- are included among the sleep disorders gence of a new clinical subspecialty con- secondary to medical problems, mainly cerned exclusively with problems of because this review does not focus heavily sleeping and waking. In any case, there on this type of problem. are a growing number of sleep disorder A third category, the parasomnias, centers in this country that attempt to includes those activities that would be use scientific knowledge of sleep and more or less normal if executed appro- sleep-wake rhythms to diagnose and priately during waking, but are patho- treat sleep disorders. (In 1972, there logical during sleep (eg, sleepwalking, were three or four centers; now, there nightmares, enuresis, bruxism). These are many more, as listed in Table II.) disorders are often called dyssomnias, but Dorland's Illustrated Medical Dictionary Primary Sleep Disorders (1975) has preempted that term by de- Primary sleep disorders are those condi- fining dyssomnia as any disorder of sleep. tions in which various abnormalities Two further terms need clarification: related to sleep are the cardinal (and Insomnia, as discussed under Sleep often the only) sign or symptom of the Needs (page 17), is defined as a chronic problem from which the patient suffers inability to obtain the amount of sleep (Williams et aI, 1974). Presumably one or necessary for efficient daytime function- more of the neurophysiological/neuro- ing. Sleep needs vary, and insomnia can- chemical mechanisms involved in regu- not be defined in terms of absolute lating the wake/sleep/REM states are hours. A person who needs nine hours. malfunctioning in these primary sleep but is chronically unable 10 obtain marc disorders.

"~7_- •• I , ' Narcolepsy: about disciplining his children.He was con- Case History: Mr. R., a 36-year-old widoweT, cerned about these spells of weakness because responded to an advertisement for research uol- they seemed to increase in frequency, but his unlerrs suffning excessive daytime sleepiness. II physician had not found anything wrung with. developed that Mr. R. had felt the need to nap his muscles and had concurred with his therapist two or three times per day sina about the age of that the clumsiness must stem from some psy- 17. HI' had been told "all his life" that he was chological problem. lazy, and nobody had ever seen anything abnor- As expected, Mr. R. showed a long REM mal other than laziness in hisfrequent naps. He period immediately upon falling asleep in the was a self-employed [armer, and his repeated lab. The rest of his lab sleep was uneventful, 10- to 15-minuie naps did not interfere to any except that Mr. R.'s sleep showed many awaken- great extent with his work. ings and excessive amounts of stage 1sleep. Both During the work-up, Mr. R. casually related Mr. R. and his physician were then advised another problem, not associated in his mind with that tlie paiient had classical narcolepsy. Sleep his frequent naps. When his children made him attacks were treated with 5 mg methylphenidate very angry and he tried to discipline them, he b.i.d., and his "muscle weakness" (cataplexy) often felt very weak in his knees and needed to yielded to 25 mg imipramine t.i.d. This regimen sit down. On several occasions, when he had produced excellent results in the beginning, but remained standing, he had collapsed on thefloor habituation to the imipramine occurred within from the "clumsiness that occasionally overcomes about six months. Mr. R. was then advised to me when I am mad at my kids." Two years earlier select periods of low work pressure when he ,- i he had sought psychotherapy for this problem, could "afford" cataplectic attacks and to refrain thinking that he must have "some hang-up" from taking imipramine during those times. In r'.

Table II Some Referral Centers for Patients with Sleep Disorders'

", Department of Psychiatry Sleep Disorders Center UAB Sleep Clinic Albany Medical College Department of Psychiatry University of Alabama Albany, NY 12208 Baylor College of Medicine Medical Center AnN: Vincenzo Castaldo, MD Houston,TX 77030 University Station AnN: Birmingham, AL 35294 Sleep Disorders Center Ismet Karacan, MD (Med) DSc AnN: Vernon Peg rem, PhD Stanford University Medical Center Sleep Evaluation Center Sleep Clinic Stanford, CA 94305 Western PSYChiatric Institute Peter Bent Brigham Hospital AnN William C Dement, MD and Clinic Boston, MA 02115 3811 O'Hara Street AnN: Quentin R. Regestein, MD Sleep Disorders Clinic Pittsburgh, PA 15261 c/o Dept. Neurosurgery AnN: David J. Kupfer, MD Sleep Disorders Evaluation Center University of Arkansas Ohio State University for Medical Sciences Sleep/Wakefulness Clinic 473 West 12th Avenue Little ROCk,AR 72201 College of Medicine and Columbus, OH 43210 AnN: Herman Flanigan, MD Dentistry AnN: Helmut S. Schmidt, MD 100 Bergen Street Dartmouth Sleep Clinic Newark, NJ 07102 Sleep-Wake Disorders Unit Dartmouth Medical School AnN: James Minard, PhD Department of Neurology Hanover, NH 03755 Montefiore Hospital and AnN: Peter Hauri, PhD Department of Clinical Medical Center Neurophysiology 111 East 21Oth Street Boston State Hospital Clarke Institute of Psychiatry Bronx, NY 10467 591 Morton Street Toronto, Canada AnN: Elliot D. Weitzman, MD " Boston, MA 02124 AnN: Harvey Moidofsky, MD AnN: Ernest Hartmann, MD University of Cincinnati Sleep Physiology Laboratory Sleep Disorders Center VA Hospital (183A) Christian R. Holmes Hospital 921 NE 13th Street ~ TnH: li~tlng dOGS net constitute ~n Eden and Bethesda Avenue enaorsement or recommendation Oklahoma City,OK 73104 Cincinnatt.OH 45219 of thOS::Q namad by thG pubJi~hQr AnN: William C. Orr, PhD AnN: Frank Zorick, MD L _ 36 this' way, potency was restored during periods of ally experience vivid dreams while still hzgher need. fully conscious. However, both sleep Classical narcolepsy is a highly specific paralysis and hypnagogic hallucinations sleep disorder. Usually, the patient suf- are less frequent than the first two symp- fers from two or more of the following toms of narcolepsy, and are not neces- four symptoms (the narcoleptic tetrad): sary to diagnose the problem. • Short, but almost irresistible daytime sleep In the laboratory, a classic narcoleptic attacks. This is the primary and most dis- patient often shows a full-blown REM abling symptom of classical narcolepsy. period occurring immediately at sleep Man y of the sleep episodes occur during onset (Rechtschaffen et al, 1963; see totally inappropriate times (eg, while Figure 11). In addition, nocturnal sleep driving or eating). Usually, these attacks is usually fragmented and poor, showing last 10 to 15 minutes, but if the patient more stage 1 and less stage 2 and delta is lying on a couch. when the attack oc- sleep than is normal. REM sleep varies curs, he may sleep for two or three hours. among narcoleptic subjects: some have In addition to the short sleep attacks, excessive amounts of REM sleep, others narcoleptic subjects usually suffer from have much less than normal. Surpris- general, chronic drowsiness. They very ingly, despite their chronic daytime rarely feel fully alert, even after pro- sleepiness, when narcoleptic subjects are longed periods in bed. asked to try to sleep as long as possible, •Cataplexy. This symptom can range they cannot sleep longer than do normal from a very transient weakness in the patients (Hishikawa et al, 1976). knees to total paralysis of all voluntary In general, the disease seems best muscles while the patient is fully con- understood as a disturbance in the bal- scious. Attacks of cataplexy are often ance between the REM system, the sleep triggered by emotions (laughing, crying, system, and the system maintaining excitement). Usually such attacks last wakefulness.Narcoleptic sleep attacks only seconds.A sudden loss of muscle during the day frequently start with an tonus is observed and is either general- intrusion of REM sleep into wakefulness. ized or limited to specific muscle groups. Cataplexy can be understood as the in- • Sleep paralysis. Either when falling appropriate triggering of that part of asleep or upon waking, the patient feels the REM system that is responsible for that he cannot move any muscles except muscle atonia during REM. Sleep paral- those controlling his, eyes. However, ysis and hypnagogic hallucinations are respiration is not affected. Sleep paraly- manifestations of REM sleep that either sis usually lasts from a few seconds to occur at sleep onset (just before waking several minutes and is very frightening. consciousness has ceased) or linger in- Patients can occasionally terminate the appropriately after awakening from attacks by first vigorously moving the REM sleep. eyes, then the eyelids, then the facial There is a fair amount of confusion in muscles, and gradually spreading the the current literature concerning the areas of movement further and further term "narcolepsy." Following the dis- away from the eyes. The paralysis is covery of Rechtschaffen et al (1963) that broken immediately if somebody touches many narcoleptic subjects had sleep- the patient. Some cases of sleep paraly- onset REM periods, some authors tried sis without any of the other narcoleptic to distinguish "classical" narcolepsy (in- symptoms have been observed, and such volving REM sleep) from "idiopathic" cases do not invariably lead later to classic narcolepsy (involving NREM mecha- narcolepsy. nisms). However, it seems conceptually •Hypnagogic hallucinations. When fall- much clearer to limit the term narcolepsy ing asleep, narcoleptic patients occasion- to the sleep disorder discussed above 37

".c.. (involving sleep-onset REM, and at least had nrue r held any job fur more than two ur one auxiliary symptom, such as cataplexy, ihrre weeks .Iince then, and he had usually been in addition to the sleep attacks), while fired fur "laziness" (jailing asleep fiVl' to ten reserving the term hypersomnia for dis- times each day). He had had numerous medical orders involving excessive sleep attacks iuork-ups during the past 20 yean, involving most medical specialties frum endocrinology to without the other aspects of narcolepsy. neurology arid psychiatry.When he was first Narcolepsy is not a rare disease. The seen at the sleep clinic he had already spent more incidence rate in the general u.s. popu- than $20,000 tofind a cause for his sleepiness. lation is estimated to be around .07% However, except for gross obesity, essential hy- (Dement et aI, 1973). It appears that this pertension, and some enlargement of the right disorder has a genetic component in ventricle, all of Mr. K:, work-ups had been many cases (Leckman and Gershon, essentiall» normal. 1976), even though the disorder often Altho~gh Mr. K. came from a relatively well- does not manifest itself until the second to-do family, he was destitute, when seen in the decade of life. However, neurological lab, because of his high medical bills and his insults in later life seem to lie at the roots inability to remain gainfully employed.His personal life was in ruins - two wives had left him of some other forms of narcolepsy (Hob- (preferring divorce to living with a chronically son, 1975). sleeping, obese, heavy snorer), and he had been The excessive daytime sleepiness and unable to mainta·in adequate social relationships .... the inappropriate napping of the narco- with friends and peers because of his excessive leptic patient can be treated with stimu- sleepiness. lants, mainly methylphenidate, pemo- In the lab, Mr. K was pleasant and polu«. lene, or dextroamphetamine, in that He fell asleep within five seconds afteT "lights order of preference. However, there is out;' but as soon as he fell asleep, his breathing some evidence that the chronic use of stopped. He wakened 35 seconds later gasping amphetamines occasionally aggravates for air. This cycle then repeated itself for the next ten hOUTSof "sleep." As soon as MT. K. showed the condition. Cataplexy, on the other signs of stage 1 sleep, he stopped breathing, then hand, usually yields either to imipramine awakened 20 to 80 seconds later gasping for or to protriptyline. The dosage and ad- air. Throughout the night, he did not sleep fur ministration of both the stimulants and more than three uninterrupted minutes.By the suppressants of cataplexy need to be morning he had totaled 562 individual arousals titrated carefully. For example, there is from sleep, and more than 75% of his "sleep" no reason to give imipramine in cases in time had been spent in sleep apneas. Visual which cataplexy is not a serious problem observation of the patient indicated that during (Zarcone, 1973; Guilleminault et aI, 1974). the periods of sleep apnea his chest heaved and No drug currently available is entirely he was straining, but no air passed through the upper airways (Figurr 12). satisfactory. Habituation may be a serious In the morning, Mr. K stated that he had had problem, as is the occasional side effect a "normal night of sleep" but that he was now of impotence in males treated with tri- more tired than he had been the previous night. cyelics. "Drug holidays" may be necessary He guessed he had awakened jive to eight times" to restore potency and to evaluate the during the night, and he was totally unaware of possibility that the chronic use of am- his very labored breathing, the heavy snoring, phetamines has started contributing to and the more than 500 sleep-apnea awakenings. the problem rather than helping it. Mr. K. was told that he suffered from upper air-UJayapneas and that the only effective treat- Sleep apneas: ment for this condition was a special form of Case History: Mr. K. was a 45-year-old, permanent tracheostomy (to be closed during the severely obese (313 lb) male with a primary com- day, but opened at night to allow breathing dur- plaint of excessive daytime sleepiness. This probing sleep). Although the entire problem and its lem started when the patient was in his early treatment were carefully explained to both Mr. 20s. Although Mr. K. seemed willing and intel- K. and his private physician, both declined to ligent, and had made ,f;(}(}d gradas in college, he consider a peTIIUl'T!l!'Tlt lrachl!uswmy for a "mere

38 Figure 77. Sleep Onset in a Normal Subject and in a Narcoleptic Subject NOle how the normal subject goes from awake to stage 1 (slow eye movements. gradual reduction of chin EMG. theta waves in the EEG) and then to stage 2 (K complexes and sle-ep spindle). The narcoleptic changes immediately from wakefulness to REM sleep (rapid eye movements. sudden decrease in chin EMG and sawtooth waves In the EEG).

Normal Subject

Fully Awake

Right eye

Slow, rolling eye movements

Left eye

EEG(C4/ear)

Narcoleptic Subject

Fully Awake

Right eye

Left eye

EEG (Cs/ear) · , '------' '------' -iera Waves Theta Waves Stage 2

Sawtooth Waves ,------.

-=- ::;;ez: . - _ :x: Figure 12.Two '~pisode$ of Sleep Apnea The patient is fully awake at point 1 (alpha waves in the EEGl. At point 2 he is starting to fall asleep Notice the slowing in his EEG (theta waves and the shallow breathing). At point, 3 the patient is fully asleep. Notice the relaxation in the chin EMG. the high-amplitude theta waves. and the absence of breathing. At point 4 the patient wakes up and breathing restarts shortly thereafter.

Right eye

Chin EMG

EEG (C./ear)

1\."

Figure 13. Nocturnal Myoclonus A stereotyped twitch in the anterior tibialis muscle. This twitch occurs about every 20-40 seconds for periods of up to three hours. The EEG indicates stage 2 sleep. The second. third. and fourth twitches cause brief awakenings.

Left anterior tibialis

Right anterior tibialis -_~_

, -, ...•

~J'v"/V"f\~"/\JV~\J'-I~.J"yAjc"~'l'V-V~"""~\N~~

l \\1 r j .J'f\y1~""'-J..!.'rJ\.Jfv,\,f\)J \,,; ",.,J'i/;-;"""~\~',\.-"'.J""r,'v"~\V~," ..'W• '",I/"\-'N'\h~~

~ "'~ #" 1':-1 , Brief Awakening

1I'";'~'M.'''''I~jl~I",''t.''. •• '''~II, ~'I1''I~I'!.~-----'*,''jj\1.I''''' to ,:: .. ;', f A /s:~·. ~. ';, ... ·Vor.J,'#1 ''''~/'V.•d " .J .,.,.;.,,J. !\;.\. J \\' ,;.-." •.J'o,.vy./''y'Ff'.N: \,.,.~,. ~ J./"V.l 'ft'",~ t'iJrr,". ,hI, ,'." J~~\~II.".i·l: -··..l.,.•}.•I'.,,/ ""r·/"';\"'''''1·,:l./-'v.J-'·'vr/~VtrJ../·'t,.i•Nr-V~'''~0.,-''''''~\ ". ~~" "/ " ".' ., ' \1yr', , ,? \ ------+------+---'-, ,~H~,(------

30 seconds ,~, 1\ \ ';

43 \ 'i<

\.\~(, I vIJ~ ,. , 1\ , ·VI'{'~, \,: l'V,I/I..!lIiI\MV~r!i.:li:J"'I,..JJ -rq

Alpha Waves .

~~j\~~.J'.)'.../v~~~~Jv~\~~

~\.W'"'~,Jl~i\!~,J""/'v---'~'-t~....v--wv-ifV~W.'vf\..,./\fj~~~"v.~"""'.'-'

Brief Awakening

",1 !. I

30 seconds

44 sleep problem:' Rather, the)1 hoped that weight associated with apnea, or they may be loss would cure the difficulty. related to the same CNS dysfunction In the four months that followed, Mr. K. be- that initially caused the apneas. Hemo- came an alcoholic, and seven months later he dynamic complications are also preva- was.caught in an attempt to steal liquor. Eleven lent (Guilleminaultet al, 1975). During months later, he died "of unknown causes" in a sleep, most apnea patients show marked state prison. elevation of blood pressure, and they

...... , Sleep apneas have been described gradually may develop essential hyper- only recently (Gastaut et al, 1965). With tension during wakefulness as well. Cor this disease, patients literally stop breath- pulmonale seems to be a possible end ing whenever they fall asleep. Sleep then result. Other complications involve the J/'--.J lightens to the point where breathing heart. According to Cuillerninault et al resumes, after 10 to 180 seconds, or the (1976), a number of severe arrhythmias patient may wake up. Patients are never are commonly observed in association aware of their sleep apneas, even when with sleep apneas, such as severe sinus they are awakened literally hundreds of arrhythmia, second-degree heart block, times each night. Episodes of repeated ventricular tachycardia, and sudden sleep apneas may last from a few minutes asystoles (lasting up to six seconds). One to a few hours, alternating with appar- might wonder how many patients with ently normal sleep, or they may take up conditions such as essential hypertension practically the entire night of a patient or cardiac problems suffer from an un- (as they did in the case of Mr. K.). The diagnosed sleep apnea syndrome. Stanford Sleep Disorder Clinic has col- The following three clinical subtypes lected a large series of sleep apnea pa- of the sleep apnea syndrome are cur- tients to study this disease in detail, and rently distinguished (Guilleminault et al, the following discussion will therefore 1976): follow their conceptualization of this a. Hypersomnia-hypoventilation syn- problem. drome - In this syndrome, upper airway Three types of sleep apneas are typi- apneas are predominant. Patients com- cally observed: plain mainly about excessive daytime Central apneas are defined by an ab- sleepiness, no matter how long they sleep sence of any respiratory effort. Although at night. They take frequent, involuntary, the upper airway seems open, no air and unrefreshing naps during the day, flow~ through it. The diaphragm stops often at totally inappropriate times. They movmg. show extreme grogginess when awaken- Upper ai'ltlJayapneas involve a collapse ing in the morning, occasionally asso- of the upper airway. Respiratory effort ciated with "sleep drunkenness" or severe persists, but there is no air exchange. headaches. Although some patients suf- Mixed apneas are a combination of the fering from this syndrome are over- two types. The apnea starts with a lack of weight and might be called "pickwickian"

;, respiratory effort. The upper airway others are of normal weight. Severe r: then seems to collapse, and a few seconds depression is a frequently encountered later ineffective respiratory efforts re- corollary of this syndrome. sume (see Figure 14). b. Insomnia-sleep apnea syndrome - Additional medical complications are Central sleep apneas seem to predomi- usually associated with severe sleep nate in this syndrome, possibly because apneas. Some complications appear to blood-gas changes are not as extreme in be secondary to the extreme respiratory this condition as they are in the hyper- effort of trying to breathe against the somnia-hypoventilation syndrome. Pa- upper airway Obstruction: others seem tients are often aware that they awaken to be the result of chronic hypoxemia frequently during the night, but they

45 I' . rarely suspect that they are having sleep alveolar hypoventilation developed and apneas. Most have taken sleeping pills, systolic arterial pressure rose and re- which invariably make matters worse. mained above the physiological level While these people are excessively tired found during wakefulness. during the day, they are rarely forced A presumptive diagnosis of sleep into taking inappropriate naps. apnea can be made in the physician's c. Sleep apnea in children - Guille- office if there is either heavy snoring or minault et al (1976) have reported on abnormal behavior during sleep, and if eight children who suffered from severe, the bed partner or a parent can docu- upper airway sleep apneas. According ment the apneic pauses. Alternatively, to these authors, excessive daytime sleep- the patient might be observed during a iness, decreased school performance, nap in the physician's office, although abnormal daytime behavior, recent patients with less-severe apneas are 0(- enuresis, morning headaches, abnormal casionaJly able to sleep for two or three weight gain, and the progressive develop- hours with regular breathing. This makes ment of hypertension all suggest the repeated observations necessary. possibility of a sleep apnea syndrome if If a presumptive diagnosis of sleep they are associated with loud and irregu- apnea syndrome is made, a work-up in lar snoring. It seems likely that "Ondine's a sleep center equipped to measure such Curse" (Severinghaus et al, 1962) is a things as blood gases and pulmonary ;.... special variant of central sleep apnea. function is mandatory. Although such Much has been said about a possible rela- work-ups are relatively expensive, costs tionship between the infant sudden death are minimal when compared to the ex- syndrome and sleep apneas (Weitzman tensive and often unrevealing medical and Graziani, 1974). Although such a work-ups most of these patients have relationship seems quite likely (Stein- undergone in the past. In addition, sleep schneider, 1972), the final data are not apneas severely damage one's health, yet in. and the risk of sudden death during The incidence of sleep apneas in the sleep is high in undiagnosed sleep apnea general population has not been deter- patients. mined. On the basis of preliminary data, It appears that upper airway sleep Dement (personal communication) esti- apneas can occasionally be based on mates that there are about 50,000 serious purely mechanical abnormalities such as sleep apnea patients in this country. excessive fat deposits, abnormally thick However, should some disorders of cur- soft palate, micrognathia, and retrog- rently unknown etiology (eg, specific nathia. Other sleep apneas are secondary nocturnal arrhythmias) be found to be to a variety of known neurological con- rooted in this syndrome, the incidence ditions (Guilleminault et al, 1975). Most of sleep apnea might be considerably often, however, upper airway apneas hig-her. seem related to a sudden, dramatic, and Practically all sleep apnea patients are apparently reflexive collapse of the heavy snorers. Thus, snoring in a patient upper airways, a finding that points to a who complains either of excessive day- specific CNS dysfunction. Similarly, in time sleepiness or of fragmented sleep central apneas, the sudden cessation of is usually the most obvious sign suggest- any respiratory effort suggests a CNS ing a sleep apnea syndrome. Indeed, dysfunction, possibly some lesion or ab- Lugaresi et al (1975) found sleep apneas normality in the lower respiratory centers. in a great number of snorers who did not - If upper airway apneas are predomi- complain of any other sleep problem. nant, potential obstructions in the airway When the apneic episodes in these heavy, might be investigated first. If the patient noncomplaining snorers were repetitive, is obese, massive weight loss has occa-

46 ~------~=~~~.-.~--.-.---~- :i sionally alleviated the syndrome (Fisher patients, but there are problems of adap- et al, 1976). In severe cases, however, tation and potential side effects. Some :1 especially when conditions such as severe children have found some relief with cardiac arrhythmias or cor pulmonale chronic phrenic nerve stimulators, but :) threaten the patient's life, a special type the results have not been impressive. 's of permanent tracheostomy seems to be Primary insomnia, primary hypersomnia r the treatment of choice and gives dra- Case History: Miss O. was a 29-year-old sec- f matic, immediate relief. The opening of retary who had apparently been an insomniac the tracheostomy is closed during the since birth. In fact, her hospital chart bore a note day for normal breathing, but opened from the newborn nursery commenting on the before sleep.Excessive daytime sleepi- remarkable lack of sleep in this baby. Miss 0.:1 ness disappears within a few days follow- earliest memories were of many preschool nights ing tracheostomy, cardiac arrhythmias when she would sit quietly by her ,window for gradually improve, and blood pressure hours while her parents slept, watching the de- returns to normal within two or three serted street below, exhausted, but unable to months. However, the medical problems sleep. She remembered grade school only as a reappear almost immediately if the constant struggle, always tired, rarely sleeping well. Although intelligent, she barely scraped by tracheostomy is plugged during sleep. in school. No adequate treatment for central Starting with high school, Miss O. noticed apneas has been found. Imipraminic that she could not take cola drinks or coffee after drugs, especially the investigational lunch, because it would keep her up "all night." drug clomipramine, have helped some Most available hypnotics and stimulants had

96.5 0, Saturation I

95.7

-10 94.6 Endoesophageal -20 Pressure -30 93.1 -40

0 -50 os 90.5 Ul q -60 C I Q) o § -70 ID Expired CO, a.. 86.5

(J) 40 I I 30 seconds E E o 20 u Mixed Periodic Respiration. a.. o Figure 14. Polygraphic Recording of a Sleep Apnea, Mixed Type

This 49-year-Old male patient is breathing normally attempts to breathe result in wide fluctuations of during the first 5 seconds. He then falls asleep (EEG endoesophageal pressure. until breathing resumes. not shown).·A central apnea of about 40 seconds' CO, is expired. and 0, saturation increases again. duration ensues (see endoesophageal pressure). On the right side of the recording, obstructed Counesyof Guilleminaull et at. West J Med 123: 7-16. July 1975.

47 been tried at some time. A lmost all had helped for down. She barely managed to finish nursing a few days or weeks, but none had brought any school, and,from then on, always requested gen- lasting relief, either from her insomnia or from eral dut-y work in overcrowded wards because her chronic, excessive daytime fatigue.Extended this kepi her constantly busy throughout her evaluations at some of this nation's leading med- eight-hour work shift. Occasionally assigned ical centers were negative, except for comments lighter duty, she would often fall asleep on the on Miss 0.:, general lack of stamina and ema- job, and this had led to her being fired on three ciated condition. Recent psychological testing different occasions. showed some excessive concern about somatic On the first laboratory night, Mrs. M. fell function and a generalized feeling of chronic asleep within two minutes after "lights out." She malaise, but no noteworthy psychopathology be- then slept for eleven consecutive hours without yond that. Specifically, no clear signs of depres- awakening, almost without body movements. In sion couLd befound. the morning, she labeled this sleep "somewhat In the lab, Miss O. slept very fitfully for three shorter than average."On another lab night, orfour hOUTSeach night, and she was very easily when she was worried about a sick child at home, aroused by even the slightest click on the inter- she slept onl» nine hours and felt excessively com. When aroused, she was immediately and tired in the morning, "like an insomniac after fully awake. Most remarkable were her very a sleepless night." No sleep apneas nocturnal "weak" sleep patterns: almost no delta waves, myoclonus, or nocturnal seizures could be re- and only very poorly defined, irregular sleep corded. Sleep was essentially normal except for spindles that occurred once every five to eight its unusual length. minutes or so. However, no respiratory or mus- Mrs. M. was placed on low doses of ampheta- cular abnormalities could be found during mines. Follow-up data one and two years later sleep, and the alternations between REM and revealed that Mrs. M. now slept about eight NREM sleep seemed normal. hours per night and felt quite awake during the , Miss O. was given sensorimotor rhythm day.The patient and her family were extremely (SMR) training, a procedure designed to in- grateful that she was now "an awake mother," crease activity in the 12 to 15 cps EEC Tange. and the required amphetamine dosage has now After about 40 hourly sessions, she seemed able been stable for at least two years.No addiction to produce somewhat more SMR. Concomitantly, developed; when Mrs. 1\11. was abruptly with- sleep at home 'improved slightly, and Miss O. was drawn from amphetamines nn two occasions, delighted. Postbiofeedback evaluations showed there were no serious withdrawal effects except about 40 minutes' m01'e sleep per night after thai excessive daytime sleepiness reemerged. biofeedback, fewer awakenings, and, possibly, On purely theoretical grounds, one some better-formed sleep spindles. However, would suspect that there must be patients sleep stili was extremely poor after biofeedback, and Mi.ss D's daytime fatigue improved only in whom the complex neurophysiologi- marginally. cal/neurochemical balance between Case History: Mrs. M: was a 52-year-old sleep and wakefulness is maladjusted in nUTSewho sought help from the lab for her exces- one way or another. Either the sleep or sive daytime sleepiness. She had suffered from the wake system might be excessively this problem since her teenage years, and she weak or excessively strong purely be- described vividly how in high school she would cause of the normal range of variability accept any active dates (such as horseback riding in any biological system or because one and dancing) but would have to avoid quiet or both of the systems might be damaged dates (such as movies) because she would fall (either anatomically or biochemically). asleep within a few minutes after sitting down. The preceding two cases appear to il- Throughout her life she had slept from 12 to L6 lustrate such problems. Miss O. shows hours per day. When her children were little, she would lie down on thefloor tofeed themJor fear primary insomnia, whereas Mrs. M. ap- of dropping them should she fall asleep during parently suffers from primary hyper- nursing. Struggling against her sleepiness, she somnia. was able to take her children bowling, swim- Although most insomniacs show psy- ming, or roller skating, but she could rarely play chiatric, medical, or behavioral etiology, cards with them or supervise their homework, a minority (about 10% to 15% at the because she would fall asleep as soon as she sat Dartmouth Sleep Clinic) do seem to

48 show demonstrably poor sleep in the effects of chronic drug usage. Others, ..' laboratory in the absence of significant however, do seem to be primary, or they medical, psychiatric, or behavioral psy- might be based on very subtle abnormal- chopathology. In some or these cases, ities such as biochemical disturbances we probably just have not learned where that are still undiagnosable with current to look for the pathology. However, it technology. Indeed, some patients at the seems likely that others in chis category Stanford Sleep Disorder Center who are showing primary insomnia. In addi- showed excessive daytime sleepiness tion to their shortened and fragmented were found to have elevated levels of sleep, such patients often show a num- 5-hydroxyindoleacetic acid (5-HIAA) in ber of atypical wave forms during sleep, their cerebral spinal fluid. This is an in- such as a lack of sleep spindling or free. teresting result, because the production quent intrusions of alpha bursts into of 5-HIAA in the brain is thought to be sleep (Phillips et al, 1975). proportional to serotonergic neuronal No direct treatment is currently avail- activity (Aghajanian et al, 1967). It is able for primary insomnia. The recently currently unknown how widespread ex- developed SMR biofeedback training, cessive CSF 5-HIAA is in hypersomnia, however, does hold some initial promise because most laboratories do not have (Feinstein et al, Jordan et ai, 1976). In the facilities to make the determination. this training, one attempts to strengthen The treatment of primary insomnia 14 cps activity over the sensorimotor and primary hypersomnia is still very cortex by biofeedback. The hope is that much in its infancy. Amphetamines are the strengthening of this rhythm will im- still the most widely used treatment for prove sleep spindling during the night, primary hypersomnia. Whether the in- as it apparently does in cats (Sterman et herent risks in such treatment are worth al, 1970). Hopefully, this will, in some as the benefit depends on the seriousness yet unexplained action, strengthen the of the hypersomnia. Fortunately, the sleep system, deepen sleep, and avoid danger of becoming seriously addicted excessive arousals. to am phetamines seems to be somewhat Similar to insomnia, most cases of ex- lower for hypersomniacs than it is for cessive sleep show medical or psychiatric people who take amphetamines for etiology, or they can be traced to side weight reduction or for "kicks." Hyper-

Table IJJThree Questions to Ask when Patients Complain of Excessive Daytime Sleepiness (EDS) ..•

Have you ever had any unusual Reason: If patients report either cataplexy or weaknesses muscular experiences? (especially knees) during excitement, suspect narcolepsy. This question helps to diagnose about 65% of all patients with EDS.

Do you snore? Reason. If "yes:' especially if snoring is irregular, investigate sleep apneas. This question helps to diagnose an additional 10% to 15% of patients with EDS.

What medications have you used Reason: Chronic use of stimulants, hypnotics, and some other chronically during the past few drugs can result in EDS. This question helps to diagnose an months or years'? additional 5% to 10% of patients with EDS.

According to Dr. W Dement, Symposium on Sleep Pathology, Boston,June 7. 1976.

49 .i Ii 'r., ". r-2~": ·:'-"""''''''''~~"""~~~""",_""a4!!!!,_;:; __ ,!,!,,!_''''''''''''''_'''''''',!,,.o !I"••••••••:III•l'I_!IO'!A_~"'. """''''. =••••e_"ffi_••_ ••IO!Eee•::""__ "'""=¥.,_"""''''''''',g:d!~,·!;i.:.i' ~~'r 'i ;' , '

; I sorn n iacs do not usually need ever- 40 seconds Usually, myoclonus is re- increasing dosages of amphetamines to corded from both the right and the left obtain the same effects, and they do not anterior tibialis muscle. Some myocloni i show excessive withdrawal effects when are bilateral; others are unilateral. Epi- Ii the amphetamines are taken away. sodes of such myocloni generally last Other medications that might be tried in from five minutes to two hours, and they selected cases are the stimulating anti- alternate with periods of normal sleep. depressants such as protriptyline Usually, periodic leg movements are not (Schmidt et aI, 1976). seen during wakefulness, and the myo- I; Nocturnal myoclonus and clonic leg twitches are never associated restless legs syndrome: with epileptic discharges in the EEG. Case History; Mr. M. was a 35-year-old elec- Nocturnal myoclonus should not be trician sent to the sleep clinic with a diagnosis of confused with the occasional "hypnic "inadequate personality" after two years of psy- jerks" or startle movements, which chiatric treatment. His main complaints were awaken most of us occasionally just as we that he[elt "washed up" in the morning and that, fall asleep. Although bed partners usu- uccasionally, he had difficulties falling asleep ally can describe nocturnal myoclonus in because his "legs were nervous." By this he meant detail, patients themselves are rarely that hefelt uncomfortable but not pamful sensa- aware of their leg movements and typi- tions creeping deep inside his calf The urge to cally complain of their fragmented and move his legs then became so strong that he had unrefreshing sleep. to gel up and waLkfor 10 to 20 minutes before he could try to sleep again. Occasionally, this Neither the incidence nor the etiology problem continued until 4 AM or 5 AM. De- of nocturnal myoclonus- has been re- tailed waking neurological evaluations and solved. Guilleminault et al (1973) postu- laboratory work-ups had been negatzve, and a late that phasic events, occurring every sleep study was performed. 20 to 40 seconds, seem improperly regu- Each night in the lab, Mr. M. got up once or lated in nocturnal myoclonus, and that twice after "lights out" to "walk off his 'legs" these events then lead to "postexcitatory W hen hi!finallyfell asleep, episodes of nocturnal rebounds" Similarly, Lugaresi et al myoclonus were recorded.For periods of up 10 (1972) relate nocturnal myoclonus to a one hour, Mr. M.\ legs jerked violently every 30 number of vegetative and somatic phe- to 40 seconds, and almost each time this hnp- pened, the EEG showed a 5- to Ti-second arousal nomena that tend to oscillate periodi- (see Figure 13, page 42). In the morning Mr.M. cally every 20 to 30 seconds (such as sys- was nut aware that he had been aroused 300 to tematic arterial pressure and peripheral 100 times by n octurn a] myoclonus, and he motor neuron excitability). knew only that he felt tired, as if he "had not At present, there is no adequate treat- slept very much." ment for nocturnal myoclonus. Diaze- Mr. M. tuas placed first on 20 mg, later on 40 pam seems to mask the problem. The mg, diazepam h.s. He reported having fewer myocloni are less violent after diazepam problems with hi. "nervous legs" before the onset and, therefore, seem to arouse the sleep- of sleet: and experiencing a more refresnmg er less often. Methysergide maleate has sleep, Objectively, when recording in the lab three months later, he did not get up after "lights been proposed as a treatment of noctur- out" to "walk off hi» legs." The episodes of noc- nal myoclonus in some cases, and some turnal myoclonus were about the same length, promising results have been reported but each discharge seemed much weaker than be- with y-aminoburyric acid analogs (Guil- fore. I-nstead of 300 to 400 limes each night, he leminault et al, 1975). now was aroused only 20 to 40 limes, and hefelt Many patients with nocturnal myoclo- better during the day. nusalso complain of thc restless legs syn- A diagnosis of nocturnal myoclonus drome (Fkbom, 1945) This syndrome is made when highly stereotyped leg occurs mainly before one falls asleep twitches repeat themselves every 20 to and at other times of inactivity while

50 awake. Uncomfortable. but 1I0l really dirappeared within a few days. A nine-montn painful, sensanons OCcur deep inside follow-up in the laboratory stunned JewRr alpha ·l thc calf (occaaionally also in [highs and waves imrudmg into his sleep, and somewhat ; feel), ~emation£ that can only bc ame- [ongRr sleep. When ii« amitriptyline W(L\ unth- liorated by moving the legs vigorously. dtuum, alpha ri'turnud, and so did lus muscular stiffness. Some': C;;tSt'S uf (he restless legs syn- dromc have been linlu:,d LO motor neu- Chronic alpha intrusion irun sleep (Hauri ron d13ca3C (Frankel ct al, 1974). Others and Hawkins, 1973) seems to be re- seem La have been secondary to defi- lated to a general feeling of somatic cicncies in iron, calcium, or vitamin E. malaise and fatigUE in the face of rela- Chronic uremia is often associated with tively adequate amounts .of sleep Re- the restless legs syndrome, as well as with cently, Moldofsky et al (1975) have nocturnal myoclonus.However, in the ma- found that many patients with fibrositis jority of cases no cause for restless legs also show chronic alpha intrusion into can be found. Boghen and Peyronnard NREM sleep. Fibrositis is a poorly de- (1976) suggest that about one third of all fined symptom complex that consists of cases involving restless legs have a famil- acute or chronic aching, soreness, and ial incidence and that this syndrome stiffness in the absence of objective ab- may be transmitted as an autosomal normalities. Moldofsky et al (1975) sug- dominant trait. gest that the observed alpha intrusion Treatment of the restless legs syn- into the sleep EEG could be related to a drome obviously involves the removal of "nonrestorative sleep syndrome" associ- any deficiency that may be found. If none ated with fibrositis. However, care seems can be found, diazepam or oxycodone indicated; the chronic use of hypnotics has been recommended, and carba- and/or stimulants may also result in an mazepine occasionally seems quite ef- excessive intrusion of alpha waves into fective in serious cases. Also, it appears sleep, without relating to fibrositis. that an adequate exercise program, cou- No treatment for nonrestorative sleep pled with some form of deep muscle has yet been validated, and not everyone relaxation (eg, meditation) during other who suffers from excessive fatigue in times of the day, has brought relief to the morning shows alpha intrusion into some restless legs syndrome sufferers. Sleep. According to Moldofsky (personal communication), neither L-tryptophan Nonrestorative sleep: nor thorazine is effective in abolishing Case History: Mr. A., a 32-year-old executive, alpha intrusion or excessive muscular Jelt that his sleep was relatively adequate in stiffness in the morning. Occasional pa- length, but very shallow.He also awakened each tients have benefited from low doses of morning with excessive JOligue and muscle stiff- . \ ness, as if he suffered from rheumatoid anhruis. sedating antidepressants, but to date the However, a work-up for arthritis proved nega- numbers are far too small to warrant tive. The muscle aching and stiffness usually definite conclusions. disappeared gradually during the morning, but "Pseudoinsomnia" Mr. A. felt somewhat tired all day, as if he had Many patients claim that they sleep not slept enough. little or not at all, even though observers When studied in the laboratory, Mr. A. slept a (ie, bed partners, nurses in hospitals) total of about seven and one half hours each find them to be soundly asleep through- night.However, throughout NREM sleep, high out the night. Such patients are usually amplitude, sloui alpha waves continuously in- truded into his sleep EEG, making it very suspected of malingering, supposedly to difficult to score the record. gain sympathy. Undoubtedly, malinger- Mr. A. was put on a regimen of 50 mg oj ing is a possible explanation in some pa- amitriptyline h.s. Subjectively, he then slept tients, and many insomniacs exagger- much more soundly, and the morning stiffness ate their sleep problems. Nevertheless,

51 subtle EEG and physiological abnormali- Neuiral-siaic syndrome: Thi3 syndrome lies have been found in a number of pa- involves excessive daytime sleepiness, tients with pseudoinsomnia, even in the usually associated with automatic be- absence of gross sleep pathology. Some havior, blackouts, and d general lack of patients, for example, show hundreds attention. Polygraphic recordings show of "mini-arousals" throughout the night, hundreds of "rnicrosleeps" (lasting 10 to each arousal lasting for only a few sec- 60 seconds each) during the day,coupled onds. Others seem to be thinking very with hundreds of "rnicrowakes" during intensely throughout the night, even the night. It appears that neither sleep when objectively they are in NREM nor wakefulness can be maintained over sleep, or they may show excessive adequate lengths of time. There is no amounts of stage 1sleep. Thus, pseudo- known treatment. insomnia should be diagnosed conserva- REM-interruption insomnia: REM sleep tively, and the daytime waking behavior is very light sleep in humans, and fre- of such patients should show personality quent awakenings from REM are com- traits commensurate with a diagnosis of monplace. It now appears that some pseudoinsomnia. If not, the patient patiems awaken chronically just after might well be given the benefit of the REM onset or even slightly before REM doubt. There are many sleep disorders is due to begin, and then have difficulties recognized today that were unknown returning to sleep. This condition is sus- just ten years ago, and others will un- pected when a patient habitually awak- doubtedly be discovered in the future. ens 1!Is!to 2 hours after sleep onset. Miscellaneous Primary Sleep Problems According to Greenberg (1967), REM- Periodic hypersomnia: Some patients interruption insomnia might be ex- sleep excessively for days, weeks, or even plained as a consequence of experienc- months, alternating with periods of nor- ing repeated, extremely unpleasant mal or even excessively short sleep. In dream episodes. Some patients showing most cases, such persons are found to be this problem might conceivably have suffering from recurrent depressive learned to avoid REM sleep because of episodes or from manic-depressive psy- these poor dreams. Greenberg has found chosis. In other patients, periodic hyper- that chlordiazepoxide (50 mg to 125 mg) somnia points to disorders such as en- was reasonably successful in inhibiting docrine problems or difficulties in the REM-interruption insomnia, whereas regulation of hormones, or a relation- diazepam (5 mg to 20 mg) had less effect. ship with menstruation (Billiard and Thus, the former medication might be Passouant, 1973; Billiard et al, 1975). tried if a patient com plains of awakening Kleine-Levin syndrome: This disorder is every 112 to 2 hours per night. However, much more rare than its exterisive cov- the unpleasant dreams, which the REM- erage in the literature indicates. In the interruption insomnia had presumably Kleine-Levin syndrome, periods of ex- attempted to avoid, might then return, cessive sleepiness are coupled with and psychotherapy might then become increased appetites for food and sex. necessary. Periods of hypersomnia alternate with Painful nocturnal erections: All males periods of short sleep and decreased ap- regularly have penile erections during petites. This condition occurs mainly in each REM episode, unless the dream young males (Critchley, 1962). The in- content is extremely anxiety-provoking creased sleep is often disturbed by ex- (Karacan, 1975; Fisher, 1965). Occasion- cessive theta waves at sleep onset and by ally, these REM erections are painful, lack of sleep spindles. No treatment hU5 awakening the sleeper. No direct treat- yet been developed. but )OllIL IJl1lic rll:~ 1111-'11i1.~:.lv'..l.il'..l.ble, but the p:ltient should Kern to outgrow th~ disease with age. be reassured that REM erections are normal and that they do not indicate any Livingston and Pauli (1974). phcnobar- abnorrn:tliry in his sexual apparatus. brtal should be tried first 10 such cases (Parenthetically, K:tr~em el aJ (1970), as of nocturnal epilep~y hUL in their view, well as others, now measure these nOL- d-arnphetamine sulph~te most effec- turnzl erecrions during REM sleep in tively treats sleep seizures. Others have patients complaining of impotence, in suggested phenytoin or carbamazepine an attempt to distinguish organic [rom for the treatment of nocturnal epilepsy. functional cases of impotence.) Sleep deprivation, as well as sleep itself, is used as an activating procedure in Sleep Disorders Secondary some EEG labs. especially in Europe to Medical Problems (Scollo-Lavizzari et al, 1975). According Neurological disorders, including epi- to Gibbs and Gibbs (1971), among pa- lepsy: Obviously, any disorder that af- tients who suffer from clinically diag- fects the balance between the neurologi- nosed epilepsy, approximately one third cal wake/sleep/REM systems (see page show seizure activity in the awake EEG 13) will manifest itself III sleep dis- recording, but in sleep and drowsiness turbances. Disorders that have been recordings. 80% to 90% show seizure documented as causing sleep abnormal- activity, ities (either directly or through a change A problem with using sleep to activate in stimulus-input levels to the wake/ seizures is that it may be "too effective; sleep/REM system) include damage to showing "abnormal" waves even in clini- the spinal cord; subcortical and cortical cally sound humans. Several spike phe- lesions; brain traumas and infections; nomena (eg, the 12 and 6 cps positive and degenerative conditions. A detailed spikes, the small sharp spikes, and the 6 discussion of these disorders is beyond cps spike and wave complexes) appear the scope of this review; the reader is in the EEG almost exclusively during referred to Williams et al (1974) for many drowsiness or sleep, but are Dot, in them- references. selves, sufficiently reliable to diagnose a It is well known that some forms of seizure disorder. According to Passouant epilepsy are aggravated by sleep. Indeed, et al (1975), the occurrence of a gener- in some relatively rare cases, epilepsy alized seizure may be facilitated by may manifest itself exclusively during NREM sleep, whereas petit mal parox- sleep. For example, in a series of 645 ysms and the brief, generalized dis- epileptic patients studied by Gibberd charges of certain myoclonic epilepsies and Bateson (1974), 38 had seizures ex- are activated by REM sleep. However, clusively during sleep. This is true for this finding has not been replicated to children as well as adults. Patients might date. then complain about poor sleep, or occa- Thyroid dysfunction: Abnormal thy- sionally they may wet their beds without roid function clearly affects sleep. Hyper- anyone being aware of their nocturnal thyroidism causes fragmented,. short seizures. Therefore, it would seem wise sleep with excessive amounts of delta to record the sleep EEGs of patients who sleep. The recovery of sleep after suc- (1) have sporadic nocturnal enuresis, (2) cessful treatment of thyrotoxicosis is have nightmares without known pre- very gradual, and it may take up to a cipitants, (3) awaken with a bitten tongue year before normal sleep patterns are or cheek, (4) have unexplained blood on achieved (Dunleavy et aI, 1974). Hypo- I• their pillow, (5) awaken with a very dis- thyroidism, on the other hand, seems to ordered bed, (6) show postictal phenom- cause excessive sleepiness and a lack of ena, (7) or are described by witnesses as delta. A return to euthyroid conditions possibly having epileptic attacks during by treatment with desiccated thyroid sleep (Boller et aI, 1975). According to gradually, but very slowly, normalizes

53 -. e;::; ,~ .&i

the amount of delta sleep (Kales et aI, ifesr t.hern selvcs in cxccssrvc daytime 1967). sleepiness. Chronic renal insufficiency: The sleep Coffee, tea, and cola drinks are now so of uremic patients is short, fragmented, common in our culture that their stimu- and disorganized (Passouant et al, 1970). lam properties are often forgotten. How- Dialysis or kidney transplants improve ever, both Brezinova et al (1975) and sleep somewhat, but do not restore it to Karacan et al (1975) have demonstrated normal levels (Karacan e t al, 1972). that coffee disturbs sleep, apparently Williams e t al (1974) speculate that even in those who claim to be unaffected uremia may be accompanied by irrever- by it. A careful medical history needs to sible neuronal changes that manifest include a query concerning these issues. themselves in abnormal sleep patterns. Some people habitually drink 20 cups of Iatrogenic problems: The fact that the coffee a day and then complain about chronic intake of hypnotics and/or stim- their serious insomnia. ulants occasionally seems to destroy sleep Miscellaneous medical problems: has been discussed on page 29. However, other medications can have similar ef- Case History: Mrs.V, a 52-year-old chronic fects. These effects are often quite atypi- insomniac, claimed that she could never sleep calor idiosynchratic, as the following more than two or three hours per night and that she had suffered from a state of chronic malaise case illustrates: for the past five years.Besides her insomnia, she Case History; Miss T., a 36-year-old nurse, also complained about vague aches and pains complained that she had developed chronic, that moved unpredictably from one part of her severe insomnia following a required three- abdomen to another. Extensive medical, neuro- month service on the night shift. Although she logical, and endocrine evaluations, done in two had returned to day work two months previously, reputable hospitals, were essentially normal. her insomnia had not yet abated as had been While being interviewed, Mrs. V. readily expected. admitted to a fair amount of marital stress. In On the assumption that Miss T. had developed addition, both psychiatric interviewing and someform of behavioral insomnia or a circadian psychological testing suggested a hypochondria- rhythm disturbance, she was treated with behav- cal reaction. The only piece that did not fit such ior therapy for a month, to no avail. In a more a diagnosis was the fact that Mrs. V actually careful interview, Miss T. revealed that about did sleep very poorly in the lab, showing two three weeks prior to her switch to night work she hours of fragmented shallow sleep on the first had been placed on an oral contraceptive, but night and four hours on the second, with no hadfailed to mention this in her initial interview. delta sleep and very little REM. Although insomnia was not a side effect listed According to numerous consultants, the basic in the Physicians' Desk Reference for this con- cause of Mrs. V5 problems seemed to be neu- traceptive, the drug was withdrawn, and Miss T. rotic. Psychotherapy was therefore recommended. started to sleep better within a week:, time. The However, no progress was made in this treat- contraceptive obviously was related to Miss T's ment, and Mrs.V soon lost confidence. She insomnia. However, it was unclear whether,In then "hopped from hospital to hospital," insisting this patient, the insomnia was related to the drug that her problem must have some medical origin. per se, or whether it was related to the psycho- A few months later, to eueryone's surprise, a logical implications of taking the medication. "median arcuate syndrome" (marked stenosis of the celiac axis secondary to hypertrophy of the Excessive stimulants: The absolute arcuate ligaments of the diaphragm) was diag- amount of food intake may have a pro- nosed. Once this stenosis was corrected, abdomi- found influence on sleep: Insomnia may nal pains gradually disappeared, sleep im- develop from starvation; excessive sleep proved, and her "hypochondriasis" vanished. from overeating (see page 27). Less obvi- The above case history demonstrates ous, but apparently quire well duclI- the FAn th at almost any medical distur- merited (Bell ct al, 1976), is the fan that bance can cause sleep problems, either cerra in common food allergies also man- directly or through the accompanying

54 ·.,;~c,,,.·-~,.~_~ .~~~"-'- ~.., - --

pain and malaise, Occasional~y, the cli~- treatment. Preferably, chis evaluation comfort is the remit of assummg a hori- should be done by a physician familiar zonta] position; in other cases, one: only with the: sleep disorders. The pathophys- becomes aware of the pain when other, iology of sleep is markedly different more intense sensations of wakefulness in many areas from the pathophysiology have diminished, Besides pain, fever has of wakefulness, encompassing difficul- also been demonstrated to fragment ties and disorders that are un known or sleep, causing many awakenings and unimportant during wakefulness or that reducing both delta sleep and REM sleep. cannot be assessed when the patient is in Karacan et al (1968) speculate that these the waking stage. sleep changes might be related to a mechanism that involves the arousal system. Sleep Disorders Secondary to Sleep-exacerbated disorders: In these Psychiatric Problems conditions, disorders do not seem to af- Classified here are disorders in which fect sleep but they become worse when a the sleep disturbance is part of a larger person falls asleep.Williams et al (1974) psychiatric problem. Distinctions -con- suggest that these disorders should be cerning what might be psychiatric and classified separately from the sleep dis- what might be medical or behavioral are orders secondary to medical problems. fairly arbitrary at present. Occasionally, They are grouped together in this review the underlying psychiatric problem is mainly for the sake of convenience. fairly well hidden, and the sleep disturb- The assumption of a horizontal pos- ance seems to be the chief complaint, as ture, even without sleep, will cause hemo- the following case history demonstrates. dynamic shifts of potential importance (eg, paroxysmal nocturnal hemoglobi- Case History: Mrs. S., the 48-year-old wife of a local businessman, had been referred to the lab nuria, Marks, 1964) as well as profound for chronic insomnia after unsuccessful trials changes in respiratory performance with many different hypnotics. At the first inter- (eg, nocturnal dyspnea and orthopnea view at the clinic, she sat in the office chair, weak related to congestive heart failure). Other from lack of sleep, with red, bloodshot eyes, but sleep-exacerbated disorders are related quite in control of herself. She responded appro- to the physiological changes that occur priately, with a polite, pleasant smile. It devel- during sleep in general or during spe- oped that she had always been a "fickle" sleeper, cific sleep stages (eg, increase and greater but "had managed until last fall." She could not variability in heart rate, respiration, and think of any reason why she should not sleep blood pressure during REM). now, and said that for thefirst ti"!-ein her life she In addition to the ones mentioned now had everything she had ever wished for and really should be happy. above, a partial list of sleep-exacerbated This last statement aroused suspicion. When disorders includes nocturnal angina and the interview stalled, Mrs. S. was given a Ror- myocardial infarctions (Karacan et al, schach test. Her answers were clearly suggestive 1969); nocturnal headaches (Dexter and of a serious but well-hidden depression, even Weitzman, 1970); nocturnal asthma though the referring physician had "ruled out" (Kales et aI, 1968); emphysema (Trask that possibility. After another hour of interview- and Cree, 1962); night cramps (Simpson, ing, she finally started to realize that for the 1969); "tired-arm syndrome" (Ford, first time in her life she had felt increasingly 1956), and many other neuromuscular useless.Her youngest son had gone to college disorders. Williams et al (1974) provide two years ago, and some neighbors who had detailed references for many of these always leaned on her for support had moved away.According to her preuiousls unconscious sleep-exacerbated problems. appraisal of her life, now verbalized for thefirst In summary, a very careful medical time, there was nothing left fOT her to do other evaluation is essential if the sleep distur- than to wait for old age and death and to bear bance is not ameliorated by customary it graciously.

55 M'Y'. 5. was hUJpitalized for a UJ()rk-nf!. Hei problems. NO[ sur prisinglv, m"ny in- sleep was recorded in the laboratory and shown sornniacs are "repressors" and arc una- 10 be seriously fragmented and curtailed. Al- ware of what is going on inside [hem- thoug]: it fluctuated widely from night to night, selves. Psychotherapy for such people Mrs. S. averaged about 35 auiakenings per differs from therapy for a patient who night, and she rarely slept for more than about presents with a chief complaint of erno- four hOUTS,often less. Much of this sleep was tional distress. Kales and Kales (1973) stage 1, with no delta sleep. have commented on the fact that in- Treatment with a sedating antidepressant somniacs need a very firm, direct ap- improved sleep in the hospital almost immediately. While undergoing therapy, MTS. S. ex- proach, not unlike the one best used in plored other ways of becoming useful again, and other "psychosomatic" problems. finally became involved in a relatively unknown According to some clinicians, psycho- charity organization for a minority group. logical stress interferes mainly with the On a follow-up nine months later, MTS. S. process of falling asleep but not with the was pronounced "remitted" by all concerned; she ensuing sleep. This is simply not true. was off all drugs and functioned quite well, both Many patients under predominantly at home and in her social life.Five nights of psychological stress fall asleep quickly, sleep in the lab, however, still revealed some apparently from pure exhaustion. They insomnia.A lthough. sleep had improved dra- . matically when compared with sleep during her then wake up often and early, as soon as hospitalization, she still stunned more awaken- their excessive psychological tension ings, more slage 1 than expected from someone and arousal overcome a still-very-strong her age, and practically no delta sleep. In addi- need to sleep. Sleep is not a stuporous tion, relatively adequate night~ fluctuated with coma; there can be a fair amount .of very poor nights in an npparently random man- mental activity during REM sleep and, ner. However, Mrs. S. was satisfied. According especially in poor sleepers, often during to her, this was the way she had slept since her NREM sleep (Rechtschaffen and Mon- early 205. roe, 1969). With all the new research on sleep Depression: Fragmented sleep is a well- disorders, emotional and psychiatric known sign of most types of unipolar problems still lie at the roots of many depression. However, it is important to sleep disturbances, It is difficult to deter- realize that some seriously depressed mine just how large this group is. Figures patients sleep quite normally (Hauri and vary from about 30% of all sleep prob- Hawkins, 1973) and others sleep for ex- lems (estimated by a neurologist-sleep cessively long times during their depres- clinician) to approximately 85% (esti- sive episodes (Michaelis and Hofmann, mated by a psychiatrist-sleep clinician). 1973). Thus, poor sleep is a frequent but Estimates are made more difficult by not a universal sign of depression, both the fact that psychopathology and poor reactive and endogenous. sleep potentiate each other. Everything Traditionally, difficulties in falling looks worse, more hopeless, and more asleep have been associated with reactive threatening after a series of sleepless or neurotic depression, whereas frequent nights. When things look that bad, it be- awakenings and early-morning insom- comes progressively more difficult to nia have been related to endogenous sleep, and a vicious circle is established. depression. Laboratory sleep studies are It should be remembered that a pa- unanimous in rejecting this claim (eg, tient who presents a chief complaint of Hawkins and Mendels, 1966; Kupfer insomnia (when there is evidence of and Foster, 1973). Although laboratory substantial emotional difficulties) has evaluations cannot objectively differen- already made an important diagnostic tiate reactive from endogenous depres- statement about himself. He has difficul- sions on the basis of awakenings, the ties seeing himself as having emotional subjective complaint of early-morning

56 ;

" awakenings is sc;{[isricallyassociated with be vastly preferable to treatment with endogenous depression (Haider, 1968). hypnotics, especially if the main dose of Enrlogcnously depressed patients feel a sed::tting antidepressant can be given wane in the morn~ng, and awakenings at bedtime. at that time may be more offensive to Manic.depressive psychods: Most pa- them than to other patients. tients suffering from bipolar depression Delta sleep is consistently curtailed in (manic-depressive syndrome) sleep most forms of depression. Short and more when they are depressed, less fractured sleep alone is not the only rea- when they are manic (Hartmann, 1968). son for this lack of delta sleep; Kupfer In addition, during depression the per- et al (1973) report low-delta sleep even centage of time spent in REM sleep is in a group of depressives who slept an often increased and during mania it is adequate amount. However, low delta decreased. However, not all bipolar pa- sleep is not diagnostic of depression as tients follow this pattern; it appears that has been claimed; it is also seen in uremia, the traditionally good sleep in the de- asthma, hypothyroidism, pregnancy, pressive phase can "flip over" into in- and many other conditions. somnia in some patients, as it appears to Contrary to early reports, REM sleep do in unipolar depression. is not curtailed in depression (Mendels Occasionally, a patient will complain and Chernik, 1972). However, a phe- of suffering from repetitive bouts with nomenon unique to serious depression insomnia and/or hypersomnia alternat- is very short REM latencies in some pa- ing with long periods of normal sleep. tients (Kupfer and Foster, 1972). It is not If such repetitive bouts cannot be associ- unusual to see certain depressed patients ated with other events (eg, seasonal aller- go into a REM period 5 to 15 minutes gies, "anniversary" depressive reactions), after the onset of sleep. Coble et al (1976) manic-depressive mechanisms might be hope that this shortened REM latency, at play much more often than is the fre- together with other sleep parameters, quently suspected, but indeed very rare, might be developed into a psychobio- Kleine-Levin syndrome (page 52). logical marker for the classification and Schizophrenia: Sleep is usually dis- treatment of affective disorders. turbed during any acute psychotic tur- Although sleep improves dramatically moil. Chronic schizophrenics, on the upon recovery from depression, most other hand, sleep surprisingly well remitted patients still sleep very poorly (Kupfer et al, 1970). In particular, the nine months after hospital discharge amount of REM sleep is within normal (Hauri et al, 1973). It could be speculated limits. This was a very disappointing that this poor sleep might index some finding, because many had hoped to very basic biochemical problem in find a key to schizophrenia in its subjec- "depression-prone" individuals. It also tive similarity to the dreaming phase of seems possible that patients with a pri- sleep. mary sleep disturbance might be at a A more careful analysis of REM sleep , , '" higher risk for depression than good does show subtle abnormalities in schiz- sleepers, possibly because they cannot ophrenic patients. When acutely ill schiz- "recharge" their batteries" during sleep ophrenic patients are REM deprived, as well as normal sleepers do. they show little or no REM rebound (Zar- In summary, a clinician will do well to cone et al, 1975). When schizophrenic search carefully for signs of depression patients are REM-deprived after remis- ! ">'

(either overt or covert) when he is con- sion of their symptoms, they seem to .!; I : ~ fronted with relatively intractable insom- show an excessive REM rebound (Zarcone nia: If depression is found or suspected, et al, 1968). Dement (1972) links these a VIgorous antidepressant regimen will findings to a possible serotonergic defect

57 in the acutely ill schizophrenic patient, differences- Some show almost continu- bur [he basic mechanisms for [his pos- ous REM sleep after abrupt withdrawal sibility are not known. from high alcohol intake (Greenberg Alcoholism: Millions of Americans and Pearlman, 1967), and others show take a glass of red wine or some other little or no REM rebound (Gross and drink before going to bed, claiming that Hastey, 1975). . it helps them sleep. What is the evidence? Even after one or two years of total Small doses of alcohol do relax tense abstinence. the sleep of many alcoholic people, helping them to fall asleep more patients remains disturbed. Delta sleep easily; however, sleep after alcohol in- is low, sleep is still excessively frag- take is disturbed (more awakenings, mented, and alcoholic patients may con- more sleep-stage changes). Alcohol also tinue to have problems falling asleep depresses REM sleep, with a compensa- (Adamson et al, 1973). It is unclear how tory increase of REM during withdrawal. much of this sleep disturbance is related Delta sleep seems to react inversely: to permanent and irreversible damage more delta during intoxication, less dur- caused by the alcoholism and how much ingwithdrawal (johnson et al, 1970; Gross of this poor sleep was there all along. et al, 1971). Depending on the dose Many insomniacs feel driven to alcohol and on the speed of alcohol metabolism, in an attempt to self-medicate their sleep a compensatory REM rebound might problems. occur either later that night or during Neurotic problems: Neurotic patients the next night (Rundell 'et al, 1972). often have severe insomnia. Occasionally Chronic alcoholic patients show a "pre- a sleep problem is the main, if not the maturely aged" sleep (joh nson et al, only, complaint of such patients. Unfor- 1970) manifested in many awakenings, tunately, nobody has extensively com- many stage changes, little or no delta pared the different types of neurotic sleep, and usually somewhat-decreased sleep problems in the laboratory. This is REM sleep. Because sleep in chronic surprising, because the various types of alcoholic patients has become frag- neurotic disorders probably constitute a mented and shallow, total time in bed is large proportion of all insomniacs. usually increased. The sleep-wake rhythm When clinically important psychiatric is blurred, and the chronic alcoholic problems are associated with insomnia, patients often show excessive daytime these should be treated first. The sleep sleepiness and take numerous naps problem will then often disappear. If it (Mello and Mendelson, 1970).During does not, it seems likely that during the heavy alcoholic binges, sleeping becomes course of the psychiatric problem some almost totally abolished, and there is the "maladaptive sleep behaviors" have possibility that delirium, the cardinal been established, which can be treated feature of delirium tremens, might be as discussed in the next section. mainly the result of prolonged, almost total sleep deprivation (Bates, 1972). Sleep Disorders Secondary to Others have speculated that delirium Behavior Problems tremens might be a breakthrough of the Behavioral factors apparently playa role REM mechanisms into wakefulness after in many sleep disorders including enu- prolonged suppression of REM during resis and depressive "escape into sleep." alcoholic bouts (Greenberg and Pearl- However, such factors have been studied man, 1967). most thoroughly with regard to in- Insomnia is a frequent, but not uni- somnia, and therefore shall be discussed versal, correlate of alcohol withdrawal. here mainly as they apply to poor sleep. Similarly, the REM rebound after alco- Behaviorally based insomnias are ob- holic bouts seems related to individual viously not as dramatic or life-threatening

58 as [hose based on medical conditions that period of his life, Mr. U. usually had lain , s such as sleep apnea:;. However, it appears awake in bed until the fighting had stopped. i· I that the large bulk of insorrmias are re- Insight-oriented psychotherapy for the pasi tuio lated to behavioral problems. In addition, years had not helped him to fall asleep morn easily. Indeed, his therapist finally fell that Mr. these behavioral insomnias, being es- U. was a "remarkably healthy" person, given the sentially based on faulty learning, are extensive famil)1 difficulties that had marred his particularly resistant to the customary childhood. Similarly, steep in the lab turned out i treatments by medication and/or psy- to be normal, except for a sleep latency of about I chotherapy. 90 minutes each night. Nevertheless, friend, }, corroborated his story that at home it usually Conditioned insomnia: took him two to three hours tofall asleep.. Case History: Mr. U. was a 19-year-old col- When Mr. U. was asked to describe his best ~, lege student who usually took two to three hours night of sleep during the past. year, he related, 1 '1: ; tofall asleep each night. He dated his problem to somewhat perplexedly, that it had occurred on I: the age of 13, when his parents finally divorced the side of a mountain. While climbing one day, I after a long period of continuous fighting, yell- he and a group of friends were overtaken b)1 ing, and occasional homicidal threats. During darkness before reaching the summit, and were

Figure 15. Performance of a Sleep-Onset Insomniac (Mr. U.) Undergoing Stimulus-Control Behavior Therapy

10-

9- 8-

4- Before During After Nine-bsontti Iwo-Yeer Behavior Therapy Behavior Therapy Behavior Follow-up Follow-up Therapy

Qj 3- Ul c o g}~ Q) :J c;;.,g 2- (no data) 2 c Q.).~ E ., :.:CJ) u~ ~ ~ 1- 0<0 ~, ~., :.:,.;,: ~~O-~,--~,~,--~,--~,--~,---,--~,---,--~,------~--- - 23456789

59 .: ,~H ~1-e.-.,---·-=I"""~~~~~~~~~:~,.~_.R.~""",,~~-~_"1_~~""~.~~~~_~~_~_;m!~~~~~"'110'""",,-~.~jl:· forced to spund thr night on a very tuirroio tedgu water beds or hammocks. These people tied to the rocks.To his utter amazement, Mr. U. frequently sleep quite well in the clinic fell asleep almost immediately and spent an because the stimuli associated with poor excellent night in this uncomfortable situation. sleep at home are not present in the lab. Mere fatigue from climbing could not have An efficient way to overcome condi- played a major role in his good sleep, because he tioned imsomnia is a procedure called was a well-trained athlete and had often tried "stimulus-control behavior therapy." to improve sleep by exercising extensively. More Basically, the goal of this therapy is to likely, it appeared that all the stimuli. surround- ing his usual sleeping arrangement at home had associate the bedroom stimuli with rapid been absent on the rocky ledge, and that this sleep onset again, rather than with frus- might have been related to his excellent sleep. tration and arousal. To achieve this, the Mr. U. was then treated with the stimulus- patient is told that he is "misusing" his control, behavior therapy to be described below. bed if he lies in it awake and frustrated. In the beginning, he had to get up eight to ten Bootzin, the behavior therapist who times per night. Nevertheless, within a five- initiated this approach (1973), recom- week period the insomnia was relieved, and Mr: mends the following set of rules: U. was able to fall asleep within 10 to 20 min- 1. Lie down in bed only when sleepy. utes on at least six out of seven nights.Follow-up data nine months later, and again two years 2. Use the bed only for sleeping; do later, showed that his insomnia had disappeared not read, watch television, or eat in bed. except for occasional stress-related poorsleep. Sexual activity is the only exception to this rule, and on such occasions the sleep How can Mr. U's inability to fall asleep instructions are to be followed after- be understood? Assume that a person wards, when you intend to sleep. who is already sleeping somewhat mar- 3. If unable to fall asleep, get up and ginally is put under severe stress (eg, the go into another room. Stay up until you death of a loved one, the divorce of one's feel really sleepy, and then return to the parents). It would then seem normal not bedroom to sleep. Get out of bed again to sleep well for a period of time because if sleep does not come easily. Remember, ofthe excessive emotional turmoil. Not in- the goal of this procedure is to associate frequently, by the time the stress is lifted, the bed with falling asleep quickly. . the whole precedure oflying down in bed, 4. If sleep does not come, repeat step turning off the light, etc., has become 3 as often as is necessary throughout the associated with (conditioned to) frustra- night. tion and sleeplessness. Even if the origi- 5. Set the alarm and get up at the same nal stress is now removed, the stimuli in time every morning regardless of how the bedroom .have been associated long much you slept during the night. This enough with frustration and tension that will help the body acquire a constant they trigger these emotions by them- sleep-wake rhythm. .selves. Occasionally reinforced by natu- 6. Do not nap during the day. rally occurring poor sleep, such a condi- If the above rules are followed, there tioned insomnia can maintain itself for will be almost no sleep during the decades, long after the original stress first night. By the second or third night, IS gone. the patient is so tired that he may fall If conditioning is an important factor asleep on the first or second attempt. in a person's insomnia, almost any change Things then fluctuate for a few weeks, in sleeping arrangements initially im- but gradually the bedroom surround- proves sleep. People with this problem ings again become associated with sleep often sleep better away from their own rather than with arousal and frustration. beds, eg, in hotels or on the couch in However, most patients need a lot of en- the living room, and they are often helped couragement during this reconditioning by unusual sleep arrangements such as period. This means at least twice-weekly

60 therapist COnE

when there is objective evidence of prog- function the next day. Some information "j ress (see Figure 15). concerning the experiments on sleep In a sample ofunselected insomniacs, need and sleep deprivation (pages 15, Bootzin (personal communication) had 18) is often helpful to such insomniacs. about a 60% success rate with this meth- Occasional "trying too hard" is diffi- od. It seems likely that a much higher cult to avoid. However, when this be- success rate could be established in a comes a habit it can cause serious insom- sample of patients preselected for signs nia. The mechanisms involved in such " of conditioned insomnia. cases are then quite similar to the ones " discussed for conditioning, except that Internal arousal (trying too hard): the association is to internal thoughts, Case History: Mrs. O.was a 23-year-old mother- not external stimuli. Assume that a per- 'r. to-be who came to the office in a state of near ex- son is not sleeping well for a'valid reason. haustion two weeks before her projected time of de- After he has been sleepless for a number livery. She claimed that she had not slept a wink of nights, sleep becomes an overriding for at least three weeks, and her appmrance supported this claim. She stated that all her life she had concern, and soon a vicious circle is es- looked forward to the natural birth of her first tablished. He desperately needs to sleep, baby and had enrolled in a Lamaze class Just as tries hard LO sleep, and therefore cannot soon 0.., she learned that she was pregnant. How- sleep. Later, the mere thought of sleep ever, recently her teacher had stated that in order may cause almost phobic reactions. to be ready for birth, the prospective mothers had The presence of internal arousal as a to gel plenty of rest and sound sleep. Then,"out oj contributing factor in insomnia is estab- the blue," this insomnia had struck Mrs. 0., and lished if the patient complains that he can she exclaimed "now the whole experience will be fall asleep easily when he does not want spoiled!" to - in lectures or in front of the TV - A IJhough sleep is usually poor in the last tri- mester of pregnancy, this was obviously not the but cannot sleep when he tries to do so. whole problem in the case of Mrs. 0. Therefore, In the laboratory, the clinician will occa- she was asked to recline on the couch and was sionally ask such people to stay awake instructed in a series of Jacobsonian relaxation for 30 minutes in order to obtain a valid •••• exercises, These exercises had been tape recorded, comparison record. So instructed, these . , and Mrs. 0.fell asleep almost beforethe tape ended. patients often fall asleep almost immedi- Mrs. 0. was given the tape for home use and ately. After 30 minutes, they are called told to perform these relaxation exercises anytime over the intercom and told that it is now she wanted to sleep. In addition, she was told time to sleep. This instruction immedi- that the deep rest that would overcome her when doing these exerciseswould be almost as good as true ately results in a dramatic increase in sleep in terms of preparing her for natural child- muscle tension and causes an alerting birth. Because an adequate performance at the response on the EEG and, occasionally, birth of her first baby was now no longer threat- a sleep latency of three to four hours. ened by her inability to sleep, she no longer tried so To date, there is no adequate treat- desperately to fall asleep. The insomnia, created ment for internal arousal. Electrornyo- by an inadvertent remark of the Lamaze teacher, graphic biofeedback has been advocated, evaporated within two nights. as have other forms of relaxation train- Sleep is one of the few things in life that ing, such as self-hypnosis and systematic cannot be improved upon by trying desensitization. A number of published h~r~er. Rather, any attempt to force sleep case reports attest to the occasional effi- ~lll inevitably result in increased activity cacy of these techniques, but solid, large- 10 the arousal system, thus abolishing. scale studies are yet to be conducted. - ••••q !Tf-:- ·--iI .

However, it has been repeatedly demon- structed to stay in bed and read; if the strated that insomniacs as a group do not problem is a conditioned association be- show greater muscle tension than do tween bed and frustration, then the good sleepers (Good, 1975) preferred therapy might be (Q sugge5t One might speculate that, if relaxation that the patient get out of the bedroom techniques work in this group, they do so and go elsewhere to read. mainly by distracting a person's attention Bedtime rituals such as brushing teeth from trying to sleep and not by their and saying prayers seem to work on a inherent ability to relax the musculature. similar principle. Good sleepers often Other methods not related to relaxation become drowsy justas soon as they com- seem potentially as effective, such as plete the routine. The close temporal the proverbial counting of sheep or the association of bedtime rituals with frus- very careful and painstaking drawing, tration and arousal in insomniacs, how- in one's own mind, of all numbers be- ever, is counterproductive, and poor tween 0 and 100 on large, black sheets of sleepers should be asked to develop imaginary paper. entirely new bedtime rituals. The matter of trying to force sleep is Disturbance of the sleep-wake rhythm: basically an attitudinal problem.There- The massive effects of irregular bed- fore, it will eventually be solved by psy- times or of staying in bed too long have chological methods involving attitudinal been discussed on pages 10-12. It is obvi- change, an area still very much in need ous that for weak sleepers regular sleep- of solid research. wake patterns are mandatory. In par- What should be done when one cannot ticularly susceptible patients, irregular sleep? If a person is comfortable and or excessively long time spent in bed may unconcerned, relaxing in bed seems result in a variety of physiological and best. However, when the insomniac endocrine disturbances that may make starts to become frustrated, angry and sleeping difficult. Therefore, disturb- upset, he should get up and engage for ances of the sleep-wake rhythm might awhile in a quiet hobby or some other well have been discussed under sleep pleasant, nonstimulating task instead of disorders secondary to medical problems. lying in bed. Ifhe continues to lie in bed They are included here mainly because while frustrated, he is in danger of de- their treatment (ie, regular and limited veloping insomnia that is conditioned to time in bed) involves behavioral methods. the bedroom stimuli. Pure admonitions to regularize time Should one read in bed until one gets spent in bed are usually as ineffective tired?The answer seems to be "yes;' if as orders to stop smoking or stop over- the reading material is not too stimulat- eating. Rather, the same techniques use- ing and if the morning arousal time re- ful for the treatment of overeating or mains fixed. Without a fixed arousal time, smoking can be used as well for regu- reading in bed will eventually lead to larizing time in bed. These techniques sleep-phase changes described on page involve such strategies as reinforcement 12. In many cases, reading a mildly pleas- of scheduling, aversive conditioning for ant but unexciting book will keep pa- oversleeping, group support, and graph- tients from trying too hard to sleep, and ing of sleep-wake behavior. drowsiness will finally overtake them. Self-image; sleep phobias: Most human It may be noted that Bootzin forbi.ds beings occasionally have poor nights of his patients to read in bed, although sleep. For good sleepers, these nights reading in bed is encouraged here. The are merely unpleasant irritations, not main difference lies in the purpose of worthy of much thought. They know that the treatment: if the problem is "trying they will sleep well soon. For persons too hard," then the patient might be in- who have labeled themselves as insom-

62 ,"Ii .jI

!lillO, however, a poor night of sleep re- to change. If one single factor that is COD- affirms this label In their minds it dern- lributing to the total insomnia can be onstrates once more that they are infe- changed, sleep often improves and this I rior, and it raises again the spectre of gives the individual renewed confidence ,I the repetition of many consecutive, sleep- in his or her ability lO sleep, thereby less nights that they have experienced. further improving sleep. Knowing how painful that would be cre- ates fear and panic in the poor sleeper, Parasomnias leading to a self-fulfilling prophecy. According to Williams and Karacan Just as the "dry" alcoholic cannot take (1974), parasomnias are defined as activi- that first drink for fear of starting another ties such as walking or urinating which, alcoholic bout, so the insomniac cannot although normal for wakefulness, create afford one sleepless night for fear of problems when performed during sleep. starting another series of sleepless nights. Most of us have experienced short Elaborate precautions are therefore periods of confusion when awakened taken, a phobia of sleepless nights devel- abruptly. Such periods of disorienta- ops, and the problem becomes aggra- tion occur most frequently when one is vated by the very means that were de- awakened from delta sleep.When par- signed to avoid it. tially aroused from such sleep, most of Phobias are often treated either with us seem quite capable of carrying out systematic desensitization or with implo- purposeful behaviors such as trips to the sive therapy. In the former, relaxation bathroom, adjusting bedcovers, closing and the presentation of a graduated windows, or turning off a bedside alarm hierarchy are used to help a patient face without later remembering these acts. ,"..,. the thought of a sleepless night without Surprisingly, somnambulism, night- fear and arousal. In implosive therapy, mares, and, to a lesser extent, enuresis, the patient is usually kept in the phobic are often related to this partial arousal situation until all signs of autonomic from delta sleep (Broughton, 1968). arousal vanish. Unfortunately, behavi- Rarely are they related to dreaming, as oral therapists have been extremely slow had been suspected. It seems that some in exploring behavioral treatment of in- especially susceptible individuals occa- somnia, and much basic research in this sionally do not rise to full consciousness area still needs to be done. when disturbed during delta sleep. The various behavioral mechanisms Rather they enter a state of confusion, discussed in this section are best under- indicated on the EEG by a mixture of stood as contributing and interacting delta (deep sleep), alpha, and beta (wak- factors, not as separate disease entities. ing) waves. Somnambulism, nightmares, Cases of "pure" conditioned insomnia and enuresis may occur in this confu- or "pure" sleep-wake rhythm disturb- sional state. ances are very rare. Rather, in almost The concept that most parasomnias Ii every case, the behavioral factors often might be manifestations of incom plete aggravate psychiatric problems (eg, arousal from delta sleep explains the fol- when conditioning aggravates the in- lowing observations: somnia of depression) or they may inter- 1. Parasomnia occurs more often in act with prImary sleep disturbances children, who have more delta sleep and, (eg, when an inherently weak sleep sys- possibly, somewhat less-established tem is combined with a pattern of irreg- boundaries between full wakefulness ular time spent in bed). In such cases, and sleep than have adults. the various factors contributing to a 2. Such activities usually occur earl)' person's sleep problem should be as- in the night, when delta sleep is most sessed, especially in terms of amenability prominent.

63 Table IV Checklist for the Work-Up of an Insomniac

Questions Actions

A.Are there possible medical reasons for the insomnia?

1. Could endocrine dysfunction, pain, thorough medical work-up neurological problems, allergies, or other problems cause the poor sleep?

2. Could a problem such as sleep apnea, interview bed partner. If positive indications are nocturnal myoclonus, or sleep epilepsy found (ie, snoring, twitches), observe sleeping be present? . patient or refer to a sleep clinic. Remember that these problems are often episodic; one normal observation does not rule them out.

3. Could the problem be related to medica- gradual withdrawal from the offending tions or to excessive amounts of stimu- medication. lants (eg, coffee, chronic use of hypnotics, or some other medications)?

.,.. . B. Are there treatable psychiatric components?

1. Does the insomniac suffer from any appropriate psychiatric treatment, both recognized psychiatric syndrome, such pharmacological and psychotherapeutic. as depression, schizophrenia, or alcoholism?

2. Is the insomniac overstressed because psychotherapy, job or marital counseling, of maladaptive personality traits such environmental interventions. as low self-esteem, inability to say "no," need for excessive power, or because of excessive environmental demands?

C. Are there treatable behavioral components? .

Is the patient either muscularly or relaxation training (ie, EMG biofeedback, psychologically tense or anxious? meditation) coupled with adequate exercise and/or tranquilizers.

2. Does the patient show signs of learned behavior therapy insomnia (ie, sleeps well anywhere but in his own bed, sleep phobias, excessive arousal when thinking about sleep).

3. Is the 24-hour rhythm disturbed attempt to regularize waking and or atypical? sleeping times.

D. What else could it be?

1. Does the problem occur every problem might be related to REM sleep 1Y2 to 2 hours?

2. Does the patient complain of poor sleep, wonder about pseudoinsomnia, tJonrestorative but his family claims the patient sleeps sleep, or hypochondria. long and well?

3. Is the patient functioning well during the patient might have a healthy hyposomnia. day, even on very little sleep?

4. Has the problem existed since birth? patient might have primary insomnia.

64 "'".L"<_C~;':"-~--=~~~~~~~~~';;'~-~~-<~"'''" «'o"--::,C=

3. Different forms of parasomnia five times per week as all these changes were made. often occur in the same person, However, six months later, sleepwalking episodes 1-. Patiems a r e difficult to aro use iucrc down to two or three per month, and now, two years laser,Mr. S. sleepwalks no more than when engaged in some form of para- two or three times a year. It appears that the ex- somnia, possibly because they are then in cessive tension from teaching resulted in some a confusional state rather than in normal psychological problem and so aroused this person sleep. that even during sleep he was highly "keyed up." 5. Because they rarely fully awaken during their parasomnias, patients rarely As discussed above, sleepwalking is recall these incidents in the morning. often a disorder of arousal, an inability to awaken completely when disturbed in delta sleep. It is not the acting out of a Somnambulism (sleepwalking): dream, and the EEG during sleepwalk- I: Case History: Mr. S., a 34-year-old math teacher, ing shows a mixture of delta and high- sought help because of his extreme difficulties in amplitude alpha waves (jacobson et al, falling asleep and because he was sleepwalking one 1965). In this confusional state, relatively to three times per week. Usually, these somnambu- simple behavior patterns can be carried listic episodes were harmless: Mr. S. got up about out, but the senses are dull; higher brain one hour after he fell asleep, rummaged around for awhile, then went back to sleep. He often slept functions are erratic, and coordination is the remainder of these nights in some unpredic poor. Reports of exquisite balancing able location, such as the living room couc~,r acts on roof tops and window ledges are the kitchen floor, or the bathtub. However, Mr. exaggerated, and somnambulists have ;1 S.'s sleepwalking episodes were not always that fallen to the-ir deaths after mistaking a benign; he occasionally urinated on the living window for a door. room carpet (apparently thinking that he was in Somnambulism 'in children is quite the bathroom), had severely hurt himself when common and of relatively little concern. stumbling overfurniture, and twice had beenfound Episodes of sleepwalking can even be trying to climb oui of a window in the couple's < fourth-story apartment. His wife was terrified of triggered in most children by lifting them these sleepwalking episodes because she couldn't to their feet during delta sleep (Kales awaken him, and she was not strong enough et al, 1966). However, there is some evi- to control him physically. dence that adult somnambulists often In the lab, M r. S. did not sleepwalk but he showed suffer from psychological problems extreme muscle tension, both during the day amenable to psychotherapy and should and when trying to sleep. He was anxious, neroous, be referred to a therapist for treatment. and claimed that he was barely keeping up with Although rare, sleepwalking may also be his teaching job and that the students were driv- associated with temporal lobe epilepsy ing him "insane."Neither tranquilizers nor sleeping pills had helped in the past.A clinical EEG or other eNS abnormalities that are and a thorough physical work-up were entirely occasionally difficult to diagnose except within normal limits. by nasopharyngeal EEG leads after For the sake of safety, the couple was advised sleep deprivation (Guilleminault, per- to move to a ground-level apartment. Mr. S. was sonal communication). enrolled in intensive psychotherapy and ex- If epilepsy seems involved, sleepwalk- changed his teaching job for a less-demanding ing may be abolished with appropriate clerical one. Furthermore, he learned deep-muscle medication (phenytoin, carbamazepine). relaxation through EMG biofeedback. Finally, In other cases, either diazepam or imi- door locks and window locks were installed in selected locations so that Mr. S. could only sleep- pramine has been reported as being walk through the bedroom, hall, and bathroom. effective against somnambulism. How- From these areas, all potentially dangerous ob- ever, many cases are not treatable by jects were removed.His wife kept the keys in her medication. The main concern in these custody in case offire. cases is the safety of a patient during his Sleepwalking episodesfirst increased tofour or walks. As was noted in the case history, this involves requiring somnambulists per week Rose awakened early in the morning, to sleep on the ground floor. removing [earju! and agitated. Although she never screamed, potentially dangerous objects before they she thrashed around in bed, moaning and groan- ing and awakening her sister who slept in the go to bed. locking windows and doors. same room. and depositing the keys with someone Rose could remember her nightmares quite well: who could open the doors in case of fire all of them dealt with strangers lurking behind or other emergency. Some somnambu- buildings, chasing her, grabbing, and attempting lists claim success by tying a rope loosely to sexually assault her. around their waist and around a bedpost. Psychological testing suggested that Rose was an When they start sleepwalking, the jerk easily alarmed and somewhat immature teenager from the rope awakens them fully, assur- starting to rebel against an excessively overprotec- ing that they will neither harm them- tive family situation. When asked what she would selves nor scare others. However, others do if episodes similar to the ones in her dreams soon learn how to untie the rope before should occur in waking life, she had no answer. She vaguely mentioned that she might scream, embarking on an episode of sleepwalking. but the men in her nightmares always went for her throat, thus making such a defense impossible. Nightmares and night terrors: Based on her psychologicallesting, Rose was re- First case history: Marc appeared to be a heal1hy ferred for short-term psychotherap».As part of this ltl-year-old who wanted to go to summer camp. treatment, the therapist suggested that she should However, his parents hesitated because three orfour always carry a strong hatpin with her and quietly times a week Marc woke up screaming, sweating, slip it into her hand should she notice suspicious and wildly flailing his arms. This usually hap- characters. Then, in reality as well as in her dreams, pened about the time his parents were getting ready she was to stab the would-be molester as soon as he to go to bed. They never could find out what terri- covered her mouth. This would at least give her time fied their son, although occasionally he mumbled to scream. something about "monsters." Marc usually calmed Rose had two more nightmares when she 'forgot" down within a few minutes and rarely remem- that she had a hatpin.As part of assertiveness bered his night terrors in the morning. The parents training, the hatpin scene was rehearsed repeatedly were mainly concerned that his bloodcurdling in tkerafrY;a week later she emerged victorious [rom screams would awaken the entire camp. They a number of her nightmares, enjoyed her victories, had also noted that Marc's night terrors increased and the nightmares disappeared. whenever he went through some new adjustment These case histories illustrate two very or stress, and they were afraid that camp might different types of frightening dreams: trigger such a period of increased night terrors. the night terror. usually associated with After extended psychological testing, the parents delta sleep (illustrated by Marc), and the were reassured that nothing seemed dramatically wrong with Marc. It was then found empirically nightmare, usually a REM dream "gone that 20 mg of imipramine h.s. suppressed Marc's astray"(illustrated by Rose). Delta night night terrors quite effectively. This was the dose terrors occur early in the night, are ac- given to Marc in camp, and he had only one night- companied by extreme physiological mare during thefour-week camp session. arousal (ie, increased heart rate, sweat- Night terrors reappeared, although with di- ing), and patients rarely remember more minished frequency, when the imipramine was with- than fragments of their content (Fisher drawn in thefall. Later, Marc's enlarged adenoids et al, 1973). Classical delta sleep night were removed for reasons entirely unrelated to his terrors appear to arise from the confu- nightmares. This, too, seemed to have a beneficial effect. Night terrors decreased to one or two per sional states. as discussed. above. They month, reappeared dramatically when he was occur more frequently when a patient transferred to a new school, but then disappeared is under stress, apparently because stress completely. causes more arousal from any form of Second case history: Rose was a high school sleep. However, the stress is usually not senior planning to go to college. However, she was directly related to the content of the night afraid that her repeated nightmares might ex- terrors, although the fragments occasion- pose her IV ridicule in the dorm. Two OT three times ally recalled in these situations obviously

66 :are fabricated by the sleeper and there- nightmares, but medication alone is fort: relate to his or her general concerns rarely appropriate. (Kahn et al, 1973). Enuresis: Bedwetting is more common Although delta sleep night terrors. than is often assumed. The incidence of are usually disorders of incomplete enuresis in apparently healthy navy re- i arousal, they may be the first clear signs cruits was 1% to 3% in both the American ! of other problems, such as temporal lobe and British navies during World War II. epilepsy or sleep apnea. Indeed, the "Primary enuresis" means that since case history of Marc is quite typical for infancy the patient has never been con- children who suffer from sleep apneas. sistently dry for at least one period of Whether or not Marc actually did suffer several months. "Secondary enuresis" 1':: from an upper airway obstruction is now means that bedwetting has reappeared '!; impossible to ascertain, because he im- after at least one dry period. According proved long before the time when sleep to Oppel et al (1968), 10% of all children • apneas were known to exist in children. at the age of seven and 3% at the age of ! i Unlike the night terrors of delta sleep, 12 are still primary enuretics. , the nightmares of REM sleep are essen- Primary enuresis, often a familial tially frightening dreams. Because REM problem, suggests organic or medical sleep is mote abundant toward morning, reasons for the enuresis, such as urethral REM nightmares usually occur in the obstructions in males and ectopic urethra early-morning hours. Physiological arous- in females. Psychological factors, on the al is much lower during REM night- other hand, are more often related to mares than during the delta night terrors, secondary enuresis, although anybody and REM nightmare sufferers usually who still often wets his bed during ado- have long and detailed recall of their lescence or adulthood should be given nightmares. a careful physical examination. There It seems important that delta night are cases where sleep studies may be nec- terrors be carefully distinguished from essary to show the relationship of noc- REM nightmares, because treatment is. turnal enuresis with nocturnal epilepsy, quite different. Delta night terrors are sleep apnea, or other sleep disturbances. often abolished by diazepam (2 mg to Although enuretic episodes can occur 5 mg for children, 5 mg to 20 mg for in any sleep stage or even when the child adults), possibly because this drug abol- is awake, many enuretic episodes (such as ishes delta sleep (Fisher et aI, 1973). night terrors and sleepwalking) begin in .. Occasionally, when the delta night terrors delta sleep and therefore occur relatively are associated with epileptic phenomena, early in the night. According to Gastaut methylphenidate becomes the treat- and Broughton (1964), enuretic episodes ment of choice. On other occasions, these usually start with a series of bladder nightmares can also be abolished by contractions during delta sleep, and eliminating the arousal stimuli (eg, Marc's they are often associated with a general .1 I swollen adenoids) that might cause dis- body movement. The sleep pattern then 'I tress and disturb delta sleep. Psycho- changes to stage 2 or 1 during micturi- therapy for delta sleep night terrors is tion, or the patient may wake up. Similar rarely indicated. to the other disorders of arousal, bed- REM nightmares are more directly wetting does increase in frequency when related to psychological conflict and the child is under stress, possibly because psychopathology in general than are psychological stress causes more arousal .', delta night terrors (Feldman and Her- during sleep and, with it, more oppor- sen, 1967; Hersen, 1971).Psychotherapy tunity for bed wetting. However, the act '1: and behavioral treatment seem to be of wetting the bed during delta sleep effective treatment methods for REM usually carries little symbolic signifi-

r., i ~ - ~~~~:$}-=->0.5 ~.%~9? _:e_. --'iiL.~~~_~"±~~~ ~~ •.;"'---~~~~~~~~ ~'l/:C#t-f .c~ .~ :::-1 ,.~i;::

cance (ic. it usually does not mean that tude is necessary, A qualified behavioral the child is particularly upset with mother therapist should be consulted before or is trying to punish the parents), On the embarking on conditioning a child with other hand, there are some rare chil- this method. . dren who consciously wet the bed while Based on the fact that bladder capacity awake, and the meaning of that action in enuretics is smaller than in normals, might be more open to psychological Starfield (1972) suggested "bladder interpretation. training." A child is asked to hold his Imipramine is quite effective in reduc- urine longer and longer in the hope that ing enuretic episodes, especially after increased bladder capacity will carry him careful titration of the dosage to the through the night (see Table V): individual child. Because patients often Nocturnal bruxism: Excessive tooth relapse when imipramine is discontin- grinding during sleep is not only annoy- ued, the drug alone is rarely the treating to the bed partner but also often ment of choice in enuresis, but its occa- leads to severe wear on the teeth. Al- sional use by both child and adult enu- though tooth grinding can occur during retics may allow normal social activity any stage of sleep, it seems related to a (eg, summer camp, visits to friends). general lightening of sleep, often occur- A number of behavioral techniques ring after an alerting response has been seem effective if enuresis persists into observed in the EEG(Reding et al, 1968). later childhood. Foremost among them To date, there is no direct treat- is the bell-and-blanket technique, where ment for tooth grinding, but some den- a bell awakens the sleeper immediately tists prescribe rubber tooth guards to be after micturition, For this technique to worn at night to prevent abrasion. Day- be successful, a very sensitive adjustment ,time bruxism has been successfully of rewards, punishments, and family atti- treated with biofeedback, and there is some evidence that such treatment may Table V Procedures Used in BladderTraining carryover into the night. Violent rhythmic movements during 1. Upon returning from school, have the child sleep: Infants, toddlers, and occasionally hold his urine as long as possible. an older child sometimes engage in violent, rhythmic head banging or body 2. To stretch the bladder, have the child drink rocking (jactatio capitis nocturna), espe- a lot of fluids while practicing retention. cially just before falling asleep.This ap- 3. After the child has held the urine as long as pears to be a normal "comfort habit," possible, collect it in a vessel marked in ounces. like thumbsucking, that is exacerbated by stress. A nine-year-old boy recently 4. Every day, record on a calendar the number studied at the Dartmouth Sleep Clinic of ounces of urine the child was able to retain. rocked more than 50% of his time in bed Indicate with a check mark if the bed was wet that morning. on the first night, both while awake and when deeply asleep. Rocking subsided 5.To divert the child's attention during the over the next few nights in the sleep training session, enlist the family in reading lab and it lasted no longer than five min- or playing with him. utes at sleep onset on the third night when he was more comfortable. 6. Once the child has attained some bladder control, encourage him to start and stop the If head banging or body rocking per- urine stream. Start this type of training early sists into later childhood, experts disagree in the morning or just before bedtime, then on whether it should be treated, and if extend to other times of the day. so, how, However, when the rhythmic

(Adapted from Starfield B: Clinical Pediatrics, movements persist into adulthood, a VOl 11, pp 343-350, June 1972). neurological consultation is indicated,

68 ,'.,I

Concluding Comments

ing in this field nor the necessary laboratory technology to properly evaluate It is hoped this review will enable physi- sleep disorders. To deal with this prob- cians to more effectively deal with most lem, the Association of Sleep Disorders sleep problems. However, in especially Centers has recently been created to intractable cases, referral to a sleep dis- establish and maintain high standards in orders center might be indicated. Al- each member clinic. Before engaging in though all-night polysomnography is ex- expensive tests, one might investigate pensive, the costs of a sleep evaluation whether a potential sleep clinic is affil- must be balanced against the costs of iated with this association. chronic, potentially inappropriate The next advances in our understand- treatments and evaluations and against ing of sleep pathology must come from

•,i· the discomfort, misery, and potential those professional and ethical clinics • health hazard of undiagnosed sleep dis- that have enough patients, clinical re- turbances. sources, and technical skills to crystal- ., In 1971, there were only three or four lize new insights into sleep pathology. sleep clinics in the United States evaluat- Unfortunately, animal models are rarely ing patients suffering specifically from applicable in this work. Man is often undiagnosed sleep disorders.Now there the only suitable subject, and this slows are more than a dozen in the U.S. alone. the pace of research and the types of Most are highly professional, ethical experimentation that can be done. Still, . ~ es.tablishments attempting to apply the in the future, we can envision a more wide range of current knowledge to complete understanding and treatment problems of human sleep. However, of highly specific sleep problems, the apparent success of sleep disorders whether through drugs, other medical centers has also attracted some so-called procedures, or individualized and spe- sleep experts who possess neither train- cific behavioral treatment.

69 " .~

Adamson J. BurdickJA: Sleep of dry alcoholics. Arch Ceo Psychiatry 28:146-149; 1973.

Aghajanian GK. Rosecrous JA, Sheard MH:Seroto- nin: Release in the forebrain by stimulation of midbrain raphe. Science 156:402-403, 1967.

Agnew HW. Webb WB. Williams RL: Comparison of stage four and I-REM sleep deprivation. Percept Mot Skills 24:851-858. 1967.

Aschoff J. Hoffman K. Pohl H. Wever R: Reentrain- rnent of circadian rhythms after phase-shifts of the zeitgeber, Chronobiologia 2:23-78, 1975.

Aserinsky E. Kleitman N: Regularly occurring periods of eye motility and concomitant phenomena during sleep. Science 118:273-274. 1953.

Baekeland F. Lasky R: Exercise and sleep patterns in college athletes. Percept Mot Skills 23:1203-1207, 1966.

Baker MA. Hayward IN: Autonomic basis for the rise in brain temperature during paradoxical sleep. Science 157:1586-1588.1967. Bates RC: Delirium tremens and sleep deprivation. Mich Med 71:941-944. 1972.

Bell IR, Rose ki nd M. Isaacs JG. et al: Provocation-neu- tralization food injection testing in EDS patients. in Chase MH. Mitler MM. Walter PL (eds): Sleep Research. Los Angeles. Brain Information Service/Brain Research Institute. UCLA, 1976, Vol 5. p 158.

Berry B, Thiessen GJ: The effects of impulsive noise on sleep. (APS-478) NRC 1159.7. Ottawa. Canada. The Bibliography Acoustics Section of the Division of Physics. National Research Council of Canada. 1970.

Billiard M. Guilleminault C, Dement WC: A menstruation-linked periodic hypersomnia. Neurology 25:436-443.1975.

Billiard M. Passouant P: Sleep study in women. in Koella WP. Levin P (eds):Sleep. First European Con- gress on Sleep Research. Basel. 1972.Basel. SKarger. 1973. pp 395-399.

Boghen D, PeyronnardJM: Myoclonus in familial restless legs syndrome. Arch Neurol 33:368-370.1976.

Boller F. Wright DG, Cavalieri R. Mitsumoto H: Paroxysmal 'Nightmares': Sequel of a stroke respon- sive to diphenylhydantoin. Neurology 25:1026-1028, 1975.

Bootzin R: Stimulus control of insomnia. Paper presented at the Symposium on the Treatment of Sleep Disorders. Am Psycholog Assoc Convention. Montreal. Canada. July 1973.

Bradley C. Meddis R: Arousal threshold in dreaming sleep. Physiological Psychology 2:109-110. 1974.

Bremer F: Cerveau isole et physiologie du sommeil. C R Sac BioI1l8:1235-124I. 1935.

Brezinova V. Oswald I: Sleep after a bedtime bever- age. Br Med J 2:431-433, 1972.

Brezinova V, Oswald I. Loudon J: Two types of insomnia: Too much waking or not enough sleep. Br J Psychiatry 126:439-445. 1975.

Broughton RJ: Sleep disorders: Disorders of arousal? Science 159;1070-1078, 1968.

70 CnntrpJl RW: Prolonged e"po

Chilce WD. Wen G: Residual performance effects of Feinsicin B, Stcrman MB, M;lCdoflald LR: Effects of .\imulnted some booms introduced during sleep. FAA. sensorimotor rhythm biofeedback training on sleep, AM.72·19 (N.tionAI Tet.hnical Information Service. in Chase MH. Stern WC, Walter PL (eds): Sleep ~p,;ngheld, VA ~~I!i 1) Department of Transportation. Rcseilrch, Vol 3. Los Angeles, Brain Information Feclcfai Aviation Admmi.I!rd!ion, OffIce of Avia tion Service/Brain Re

Crisp AH, Stonehill E, Fenton GW, Fenwick PBC: Fisher C, Gross J, Zuch J: Cycle of penile erection syn- Sleep patterns in obese patients during weight reduc- chronous with dreaming (REM) sleep. Arch Gen Psy- tion. Psych other Psychosom 22: 159-165,1973. chiatry 12:29-45, 1965.

Critchley M: Periodic hyposomnia and megaphagia in Fisher C, Kahn E, Edwards A, et aI: A psychophysiological adolescent males. Brain 85:62.7-656, 1962. study of nightmares and night terrors: The suppression of stage 4 night terrors with diazepam. Arch Gen Psy- Dement WC: Dream recall and eye movement during chiatry 28:252·259, 1973. . sleep in schizophrenics and normals. J Nerv Ment Dis 122:263·269,1955. Flemenbaum A: Pavor nocturnus:A complication of single Dement WC: Some Must Watch While Some Must daily tricyclic or neuroleptic dosage. Am J Psychiatry Sleep. San Francisco, WHFreeman & Co, 1974. 133:570·572,1976. Dement WC, Carskadon M, Ley R: The prevalence of Ford FR: The tired arm syndrome. A common condi- narcolepsy II, in Chase MH, Stern WC, Walter PL (eds): (ion manifest by nocturnal pain in the arm and numb- Sleep Research.Los Angeles, Brain Information ness of the hand. Bull Johns Hopkins Hospital 98: Service/Brain Research Institute,UCLA, 1973, Vol 2, 464·466, 1956. P 47. Foulkes WD: Dream reports from different stages of Dement WC, Kales A, Zarcone V, et al: Effects of the sleep: J Abnorm Psycho I 65:14·25, 1962. fluidized bed on sleep patterns of normal subjects. Foulkes D, Vogel G: Mental activity at sleep onset. J Paper presented at the First International Congress Abnonn Psychol 70:231·243, 1965. of the Association for the Psychophysiological Study of Sleep, Bruges,Belgium, June 19·23, 1971. Frankel BL, Patten BM, Gillin ]C: Restless legs syn· drome. Sleep-electroencephalographic and neurologic Dement W, Kleitman N: Cyclic variations in EEG duro findings. JAMA 230:1302·1303, 1974. ing sleep and their relation to eye movements, body motility, and dreaming. ElectroencephalogrClin Franklin LM: Sleep and hypnotics in a psychiatric ad- Neurophysiol 9:673·690, 1957. mission ward.N Z MedJ 69:353·355, 1969.

Dement W, Zarcone V, Ferguson J, et al: Some parallel Casraut H, Broughton R: A clinical study of episodic findings in schizophrenic patients and serotonin- phenomena during sleep, in Wortis J (ed): Recent depleted cats, in Siva Sankar DV (ed): Schizophrenia Advances in Biological Psychiatry. The proceedings Current Concepts and Research. Hicksville, NY, PJD of the 19th Annual Convention and Scientific Program Publications Ltd, 1969, pp 775·811. of the Society of Biological Psychiatry, May 1·3, 1964. Dexter JD, Weitzman ED: The relationship of noctur- New York, Plenum, Vol 7, 1964, pp 197·221. nal headaches to sleep stage patterns. Neurology 20: Castaut H, Tassinari C, Duron B: Etude polygraphique 513·518,1970. des manifestations episodique (hypniques et respira- Dobbs ME: Beh,!Y.!.oIaLLesponses_during sleep of men toires) diurnes et nocturnes du syndrome de Pickwick. and women foaircraft noises. Paper presented at the Rev Neural (Paris) 112:573·579,1965. 12th Annual Meeting of the Association for the Psycho- logical Study of Sleep, Lake Minnewaska, New York, Gibberd FB, Bateson MC: Sleep epilepsy: Its pattern May 4· 7, 1972. and prognosis. Br Med J 2:403·405, 1974.

Dunleavy DLF, Oswald I, Brown P, er al: Hyperthyroid. Gibbs FA, Gibbs EL: How much do sleep recordings IS':', sleep, and growth hormone. Electroencephalogr contribute to the detection of seizure activity? Clin ClIo Neurophysiol 36:259·263, 1974. Electroencephalography 2:169·172,1971.

Ekbom KA: Restless legs.Acta Med Scand 158:1. Globus G, Friedmann], Cohen H, et al: The effects of 123,1945. aircraft noise on sleep electrophysiology as recorded

71 in me nume, in Wdld WD (cd): Pro~"oding~ of the Hartmann E: Longitudinal studies of sleep and dream Interno.tional Congress on Noiee 110 a Public H"o.lth paltcrn, in manic-depressive ·patienrs. Arch Gcn Problem. WashIngton. DC. U.5. Euvironrucnrel Pro- P'ychiarry 19:312-329. 1968. tccrion Agency. 1974, pp 5S7_59j Hartmann E: Pharmacological studies of elcep and Good R: Frontalis muscle tcnsi on and sleep latency. dreaming - chemical and clinical relationahips. BioI Psychophjaiology 12:465-467, 1975. Psychiatry 1:243-258, 1969. Hartmann E: Sleep requtremeru: Long alccpera, short Greenberg R; Dream inter-ruption insomnia. J Ncrv Ment Dis 144:18-21, 1967. sleepers. variable sleepers, and insorn niacs. Psychoso- matics 14:95-103, 1973. Greenberg R, Pearlman C: Delirium tremens and Hartmann E, Cravens J, List S: Hypnotic effects of L- dreaming. Am] Psychiatry 124:37-46, 1967. tryptophan. Arch Geo Psychiatry 31:394-397, 1974. Greenblatt DJ. Miller RR: Rational use of psycho- Hauri P: Effects of evening activity on early night tropic drugs.] Maine Med Assoc 65:192-197,1974. sleep. Psychophysiology 4:267-277,1968. Gross MM, Goodenough DR, Hastey JM, et al: Sleep Hauri P: The influence of evening activity on the disturbances in alcoholic intoxication and withdrawal, onset of sleep. Psychophysiology 5:426-430, 1969. in Mello N K, Mendelson J H (eds): Recent Advances in Studies of Alcoholism. U.S. Department of Health, Hauri P: Evening activity, sleep mentation, and subjec- Education and Welfare, Health Services and Mental tive sleep quality. ] Aboorm Psychol 76:270-275, Health Administration, National Institute of Mental 1970. Health, National Institute on Alcohol Abuse and Alcoholism, 1971, pp 317-397. Hauri P: Sleep in depression. Psychiatr Annals 4: 45-62,1974. Gross MM, Hastey JM: A note on REM rebound during experimental alcohol withdrawal in alcoholics, in Hauri P, Chernik D, Hawkins D, et al:Sleep of de- Gross MM (ed): Advances in Experimental Medicine pressed patients in remission. Arch Gen Psychiatry and Biology. New York, Plenum Press, Vol 59, 1975, 31:386-391,1974. pp 509-513. Hauri P, Hawkins DR: Individual differences in the Guilleminault C, Billiard M, Momplaisir J, et al: Altered sleep of depression, in Jovanovic UJ (ed): The Nature states of consciousness in disorders of daytime sleepi- of Sleep. Stuttgart, Gustav Fischer, 1973, pp 193-rn ness. J Neurol Sci 26:377-393, 1975. Hauri P, Hawkins DR: Alpha-delta sleep. Electroen- cephalogr Clin Neurophysiol 34:233-237, 1973. Guilleminault C,Carskadon M, Dement WC: On the treatment of rapid eye movement narcolepsy.Arch Hawkins DR, Mendels J: Sleep disturbance in depres- Neurol 30:90-93, 1974. sive syndromes. Am] Psychiatry 123:682-690, 1966.

Guilleminault C, Eldridge FL, Simmon FB, et al: Sleep Herbert M, Wilkinson RT: The effects of noise-dis- apnea syndrome: Can it induce hemodynamic changes? turbed sleep on subsequent performance, in Pro- West] Med 123:7-16,1975. ceedings of the International Congress on Noise as a Public Health Problem, Dubrovnik, Yugoslavia; May Guilleminault C, Eldridge FL, Simmons FB: Sleep 13-18.1973. apnea in eight children. Pediatrics 58:23-30, 1976. Hersen M: Personality characteristics of nightmare Guilleminault C, Henriksen S, Wilson R, et al: Nocturnal sufferers.] Nerv Men! Dis 153:27-31, 1971. myoclonus and phasic events, in Chase MH, Stern WC, Walter PL (eds): Sleep Research. Los Angeles, Brain Hess WR: Symptomatik des durch Elektrischen Reiz Information Service/Brain Research Institute, UCLA, ausgelosten Schlafes und die Topographie des Schlaf- 1973, Vol 2, P 151. zentrums. Helv Physiolog Pharmacolog I, C61, 1943.

Guilleminault C, Raynal D, Phillips R, et al: Action of Hess WR: Diencephalon-autonomic and extrapyra- GABA derivative on sleep patients with nocturnal midal functions. New York, Grnne and Stratton, 1954. myoclonus and idiopathic insomnia, in Chase MH, Hishikawa Y,Wakamatsu H, Furuya E, et al:Sleep satia- Stern Walter PL (eds): Sleep Research. Los we, tion in narcoleptic patients. Electroencephalogr Clin Angeles, Brain Information Service/Brain Research Neurophysiol 41:1-18, 1976. Institute, UCLA, 1975, Vol 4, ~219. Hobson JA: The cellular basis of sleep cycle control. Guilleminault C, Ttlkian A, Lehrman K, Dement WC: Adv Sleep Res 1:217-250, 1974. Cardiac arrhythmias and sleep apnea, in Chase MH, Mirier MM, Walter PL (eds): Sleep Research. Los Hobson JA: Sleep after exercise. Science 162:1503-1505, Angeles, Brain Information Service/Brain Research 1968. Institute, UCLA. Vol 5, 1976, P 172. Hobson JA: Dreaming sleep attacks and desynchronized sleep enhancement. Arch Gen Psychiatry 32:1421- Guilleminault C, Tilkian A, Dement WC:The sleep 1424,1975. apnea syndromes. Annu Rev Med 27:465-484, 1976. Hobson JA, McCarley RW, Wyzinski PW: Sleep cycle Haider I: Patterns of insomnia in depressive illness: A oscillation: Reciprocal discharge by two brainstem neuro- subjective evaluation. Br] Psychiatry 1\4:1127- nal groups.Science 189:55-58, 1975. 1132, 1968. Home JA, Porter JM: Time of day effects with standard- Haider I, Oswald I: Late brain recovery processes after ized exercise upon subsequent sleep. Electroencephalogr drug overdose. Br Med] 2:318-322, 1970. Clin Neurophysiol 40:178-184, 1976.

72 I J.cob< HI.. McGinty DJ: Effects of food deprivation on Kales A. Kales J: Recent advances in the diagnosis and sleep and wzkefulncaa In the r-at. E"p Neurol 30:212- treatment of sleep disorders, in Usdin G (ed): Sleep Research and Clinical Practice. New York, Brunnerl I 222,l97L Mazel, 1973. pp 59-94. Jacobson A, Kales A, Lehmann D. et al: So~namb~li5l1l: Kales A, Kales JD: Sleep disorders. Recent findings in I All-night electroencephalographIc atud ies. SCIence 144:975-977, 1965. the diagnosis and treatment of distu rbed sleep. N Engl J Med 290: 487-499, 1974. Johns MW: Sleep and hypnotic drugs. Drugs 9:448-478, 1975. Kales A, Kales JD, Bixler EO, et al: Effectiveness of hypnotic drugs with prolonged use: Flurazeparn and Johnson LC: Psychological and physiological changes pentobarbital. Clin Pharmacol Ther 18:356-363, 1975. following total sleep deprivation, in Kales AA (ed): Sleep _ Physiology &: Pathology. Philadelphia, Lippincott, Kales A, Tan TL, Kollar EJ, et al: Sleep patterns fol- 1969, pp 206-220. lowing 205 hours of sleep deprivation. Psychosom Med 32:189-200, 1970. Johnson LC: Are stages of sleep related to waking behavior? Am Sci 61:326-338, 1973. Kales J, Tan TC, Swearingen C, et al: Are over-the- counter sleep medications effective? All-night EEG Johnson LC, Burdick JA, Smith J: Sleep during alcohol studies. CurrTher Res 13:143-151, 197L intake and withdrawal in the chronic alcoholic. Arch Gen Psychiatry 22:406-418, 1970. Karacan I, Blackburn AB, Thornby JI, et al: The effect of doxepin H CI (Sinequan) on sleep patterns and Johnson LC, Wilson MR: Habituation during sleeping clinical symptomatology of neurotic depressed patients and waking. Psychophysiology 12:574-584, 19.75. with sleep disturbance, in Mendels J (ed): Sinequan Jones HS, Oswald I: Two cases of healthy insomnia. (doxepin HCI): A Monograph of Recent Clinical Electroencephalogr Clin Neurophysiol 24:378-380, Studies. Amsterdam, Excerpta Medica. 1975, pp 4-22. 1968. Karacan I, Thornby JI, Anch M. et al: Prevalence of Jordan JB, Hauri P, Phelps PJ: The sensorimotor sleep disturbance in a primarily urban Florida county. rhythm (SMR) in insomnia, in Chase MH, Mitler MM, Soc Sci Med 10:239-244, 1976. Walter PL (eds): Sleep Research, Los Angeles, Brain Karacan I, Thornby JI, Booth GH, et al: Dose-response Information Service/Brain Research Institute, UCLA, effects of coffee on objective (EEG) and subjective 1976, Vol 5, P 175. measures of sleep, in Levin P, Koella WP (eds): Sleep Jouvet M: Recherches sur les structures nervouses et 1974. Second European Congress on Sleep Research, mechanismes responsibles des differents phases du Rome 1974. Basel, SKarger, 1975, pp 504-509. sommeil physiologique. Arch Ital Bioi 100:125-206, 1962. Karacan I, Williams RL, Bose J, et al: Insomnia in hemodialytic and kidney transplant patients (ab- Jouvel M: Paradoxical sleep.A study of its nature and stracted). Psychophysiology 9:137, 1972. mechanisms, in Himwich WA, Schade JP (eds): Sleep Mechanisms: Progress in Brain Research. Amster- Karacan I, Williams RL, Salis PJ: Nocturnal penile dam, Elsevier Publishing Co, Vol 20, 1965. tumescence and the genesis of male impotence. in Chase MH, Mitler MM, Walter PL (eds): Sleep Re- Jouvet M, Michel F: Correlations electromyograph- search. Los Angeles, Brain Information Service/Brain Igue due sommeil chex le chat decortigue et mesen- Research Institute, UCLA, 1975, Vol 5. p 178. cephaligue chronique. C R Soc Bioi 153:422-425.1959. Karacan I, Williams RL. Salis Pl. et al: New approaches Kahn E, Fisher C, Edwards A. et aI: Mental content of to the evaluation and treatment of insomnia (pre- . , stage 4 night terrors. Proceedings, 81st Annual Con- liminary results). Psychosomatics 12:81-88, 197L vention, American Psychological Association, 1973, pp 499-500. Karacan I, Williams RL: The relationship of sleep disturbances to psychopathology. in Hartmann E (ed): Kales A, Beall GN, Bajor GF, et al: Sleep studies in asth- Sleep and Dreaming, (International Psychiatry matic adults: Relationship of attacks to sleep stage and Clinics, Vol 7, No.2). Boston, Little Brown, 1970, time of night. J Allergy 41: 164-173. 1968. pp 93-IIL Kales A. Bixler EO. Tan TL. et al: Chronic hyp- Karacan I, Williams RL, Taylor WJ: Sleep characteris- netic-drug use. Ineffectiveness, drug-withdrawal in- tics of patients with angina pectoris. Psychosomatics sornrna, and dependence. JAMA 227:513-517, 1974. 10:280-284,1969.

Kales AA, Hauri P, Bixler EO. et al: Effectiveness of Karacan I, Williams R, Thornby JI, et al: Sleep-related secobarbital with intermediate term use. Clin Phar- penile tumescence as a function of age. Am J Psychia- macol Ther 20:541-545, 1976. try 132:932-937, 1975.

Kales A, Heuser G, Jacobson A, et al: All night sleep Karacan I. Wolff SM, Williams RL, et al: The effects studies in hypothyroid patients. before and after treat- of fever on sleep and dream patterns. Psychoso- rneru. J Clin Endocrinol Metab 27:1593-1599, 1967. matics 9:331-339,1968. ~ales A,Jacobson A. Paulson M, et al: Somnambulism: Kinkel HJ, Maxion H: Schla£physiologische Unter- sychophysiological correlates. Arch Gen Psychiatry suchungen zur Beurteilung verschiedener Matratzen. 14:586-594,1966. Int Zeit Angewand Physiolog 28:247-262, 1970. Kales A, Kales J: Evaluation. diagnosis, and treatment Kleitman E, Mullin FJ, Cooperman NR, et al: Sleep f °2 clinical conditions related to sleep.jAMA 213:2229- Characteristics.Chicago, University of Chicago 235. 1970. Press, 1937.

73 1~~~~~-~~~'=--"~- --~o..-~~-==-"''''-'''Li''''"'","",~ ~J

Kleitrnan N: Sleep and Wakefulness. Chicago. Uni- Meddis R, Pearson AJD. Langford G: An extreme case versity of Chicago Press, 1963. of healthy insomnia. Electroencephalogr Clin Neu- rophysiol 35:213-214,1973. Kline MV, Sullivan PA, Coleman LL: Some clinical sleep parameters with the innerspace flotation bed:A Mello NK, Mendelson J H: Behavioral studies of sleep preliminary report with reference [Q insomnia. ] Am patterns in alcoholics during intoxication and with- Sac Psychosom Dent Med 21:3-9, 1974. drawaLJ Pharmacol Exp Ther 175:94-112, 1970. Kolata GB:Brain biochemistry: Effects of diet. Science Mendels J. Chernik DA: REM sleep and depression, 192:41-42, 1976. in Chase MH, Stern WC, Walter DL (eds): Sleep Re- search. Los Angeles, Brain Information Service/ Kupfer Dj, Foster GF: Interval between onset of sleep Brain Research Institute, UCLA, 1972, Vol I, P 141. and rapid-eye-movement sleep as an indicator of depression. Lancet 2:684-686, September'l972. Michaelis R, Hofmann E:Zur Phanomenologie und Atiopathogenese der Hypersomnie bei Endogen- Kupfer Dj, Foster FG: Sleep and activity in a psychotic phasischen Depressionen, in Kovanovic UJ (ed): The depression. ] Nerv Ment Dis 156:341-348,1973. Nature of Sleep. Stuttgart. Gustav Fischer, 1973, pp Kupfer DJ. Foster FG, Detre TP: Sleep continuity 190-193. changes in depression. Dis Nerv Syst 34:192-195, Moldofsky H, Scarisbrick P: Induction of neuras- 1973. thenic musculoskeletal pain syndrome by selective Kupfer DJ, Wyatt Rj, Scott ] , et al: Sleep disturbance sleep stage deprivation. Psychosom Med 38:35-44, -~ in acute schizophrenic patients. Am] Psychiatry 126: 1976. 1213-1223,1970. Moldofsky H,Scarisbrick P, England R, et al: Muscu- LaceyjH, Crisp AH, Kalucy RS, et al: Weight gain and loskeletal symptoms and non-REM sleep disturbance the sleeping electroencephalogram: Study of 10 in patients with "fibrositis syndrome" and healthy sub- patients with anorexia nervosa. Br Med ] 4:556-558, jects. Psychosom Med 37:341-351, 1975. 1975. Monnier M, Halt AM,Cueni LB, et al: Humoral transmission of sleep. Pflugers Arch 331:257-265, Leckman jF, Gershon ES: A genetic model of narcolepsy. BrJ Psychiatry 128:276-279, 1976. 1972. Monroe LJ: Transient changes in EEG sleep patterns LeVere TE, Bartus RT, Hart FD: Electroencephalo- of married good sleepers: The effects of altering sleep- graphic and behavioral effects of nocturnally occur- ing arrangement. Psychophysiology 6:330-337, 1969. ring jet aircraft sounds. Aerospace Med 43:384-389. 1972. Morgan BB: Effects of continuous work and sleep loss in the reduction and recovery of work efficiency.Am Livingston S, Pauli LL: Sleep epilepsy. Br Med ] 4: Ind Hyg Assoc ] 13-20, 1974. 104,1974. Morgan PA: Effects of noise upon sleep. Institute of Lugaresi E, Coccagna G,Farneti P. et al: Snoring. Sound and Vibration Research,University of South- Electroencephalogr Clin Neurophysiol 39:59-64, hampton Operational Acoustics and Audiology Group. 1975. National Technical Information Service, Springfield Lugaresi E, Coccagna G, Mantovani M, et al: Some VA22150, 1970. periodic phenomena arising during drowsiness and Moruzzi G. Magoun HW: Brain stem reticular forma- sleep in man. Electroencephalogr Clin Neurophys- tion and activation of the EEG. Electroencephalogr iol 32:701-705, 1972. Clin Neurophysiol 1: 455-473,1949. Lukas jS, Kryter KD: Awakening effects of simulated Mullaney DJ,Johnson LC, Naitoh P:Sleep before and sonic booms and subsonic aircraft noise on six sub- after gradual sleep reduction, in Chase MH, Mider MM, jects,7 to 72 years of age. NASA CR-1599.Stanford Walter PL (eds): Sleep Research. Los Angeles, Brain Research Institute, 1970, pp I-56. Information Service/Brain Research Institute, UCLA, Lund R: Personality factors and desynchronization Vol 5, 1976,P 193. of circadian rhythms. Psychosom Med 36:224-228, Nauta WJH: Hypothalamic regulation of sleep in rats. 1974. ] Neurophysiol 9:285-316, 1946. MacFadyen UM, Oswald I, Lewis SA:Starvation and Oppel WC, Harper PA, Rido RV:The age of attaining human slow-wave sleep. ] Appl Physiol 35:391-394, 1973. bladder control. Pediatrics 42:614-626, 1968. Oswald I: Drugs and sleep. Pharmacol Rev 20:273- Marks j: The rnarchiafava micheli syndrome (pa- 303,1968. roxysmal nocturnal haemoglobinuria). Q] Med 18: I05-l2I,1949. Oswald I: Sleep and its disorders. Handbook Clin Neu- roI3:80-1I1,1969. Mauthner L: Zur Pathologie and Physiologie des Schlafes. Wien Med Wachenschr 40:445-446, 1890_ Othmer E: Persiinlichkeit and Schlafverhalten. Psy- chologia Universalis 9:1-128, 1965. McDonald DG, Carpenter FA: Habituation of the ori- enting response in sleep. Psychophyoiology 12:618- Ouo E: Physiological analysis of sleep disturbances in- 623,1975. duced by noise and increased room temperature, in Koella WP, Levin P (eds): Sleep. First European Con- MCFarland RA: Air travel aCTOSStime zones. Am Sci gress on Sleep Research, Basel, 1972. Basel, S Karger. 63:23-30,1975. 1973, pp 414-418.

74 Otto E, Kr~m~r H. Brauer D: Einfluss erh ohrcr Raum- Rosekind M. Phillips P, Rappaport J .• , al: Effects of lulttempcratur auf Herz,chlagfrcqucnz, Bewegungs- warcrbcd surface on sleep: A pilot study, in Chase M H, haufigkeil, Rectzhcmpcratur und Elektrocnccphalo- Mitler MM, Walter PL (cds): Sleep Research. Los c-rarnrn schlafender Menschen. Int Arch Arbcitsmed Angeles, Brain Information Service/Brain Research 28:189-202.1971. Institute, UCLA. 1976. Vol 5, p 132.

Pappcnheimer JR:The sleep factor. Sci Am 235:24- Roth T, Kramer M. Trinder J: The effect of noise 29,1976. during sleep on the sleep patterns of different age groups. Can Psychiatr.Assoc 17:197-201, 1972. Parmeggiani PL, -Rabin:C: Sleep and environmental J temperature. Arch Ital Bioi 108:369-387,1970. Rundell OH, Lester BK. Griffiths WJ. et al:Alcohol and sleep in young adults. Psychopharmacologia 26: Passouant P, Besset A. Carriere A, et al: Night sleep 201-218,1972. and generalized epilepsies, in Levin P, Koella WP (eds): Sleep. Second European Congress on Sleep Research, Ryback RS, Trimble RW, Lewis OF. e t al: Psychobio- Rome, 1974. Basel, S Karger, 1975, pp 185-196. logic effects of prolonged weightlessness (bed rest) in young healthy volunteers. Aerospace Med 42:408-415, Passouant P, CadilhacJ, Baldy-Moulinier M, et al: Etude 1971. du sommeil nocturne chez des uremiques chroniques soumis a une epuration exrrarenale. Electroence- Ryback RS. Lewis OF, Lessard CS: Psychobiologic phalogr Clin Neurophysiol 29:441-449,1970. effects of prolonged bed rest (weightlessness) in young, healthy volunteers (Study II). Aerospace Med Pflug B: Uber den Schlafentzug in der Ambulanten 42:529-535,1971. Therapie Endogener Depression. Nervenarzt 43:614- 622,1972. Schmidt HS, Clark RW, Hyman PR: Prorriptyline: Effective treatment for disorders of inappropriate or Phillips RL, Spiegel R,Clayton D, et al: A study of short excessive sleep, in Chase MH, Mitler MM. Walter PL arousals in insomniacs and normals, in Chase MH, (eds): Sleep Research.Los Angeles. Brain Information Stern WC, Walter PL(eds): Sleep Research.Los Angeles, ServicelBrain Research Institute. UCLA, 1976, Vol 5. Brain Information Service/Brain Research Institute, P 187. UCLA. 1975, Vol 4, P 231. Schmidt-Kessen W, Kendel K: Einfluss der Raurntern- Presley JM, Ellen P, Foshee DP: Environmental temp- peratur auf den Nachtschlaf. Res Exp Med (Berlin) erature and sleep in psychiatric patients. Newsletter for 160:220-233. 1973. Research in Mental Health and Behavioral Science, IB IS-IS vol 15:17-19, Nov 1973. Schulte W: Der Schlafenrzug und seine Folgen. Med Klinik 54:969, J959. Raboutet J, Lesev r e N, Re mond A: Involuntary sleeping during EEG, research on promoting factors. Scollo-Lavizzari G, Pralle W, De la Cruz N: Activation Revue Neurologique 101:404-408, 1959. effects of sleep deprivation and sleep in seizure patients: An electroencephalographic study. Eur Rechtschaffen A: Polygraphic aspects of insomnia, in Neurol 13:1-5, 1975. Castaut H, Lugaresi E, Berticeroni G, er al (eds):The Scott TO: The effects of continuous, high intensity, abnormalities of sleep in man: Proceedings of the white noise on the human sleep cycle. Psychophysi- 15th European Meeting on Electroencephalography. ology 9:227-232,1972. Bologna. A. Gaggi, 1968, pp 109-128. Sevcringhaus J, Mitchell P: Failure of respiratory Rechtschaffen A. Foulkes D: Effect or visu al stimuli on center aurornarically while awake. Clio Res 10: 122. dream content. Percept Mot Skills 20:1149-1160, 1962. . \ 1965. Shapiro CM, Moore AT, Mitchell 0 et al: How well Rechtschaffen A. Hauri P, Zeitlin M: Auditory awaken- does man thermoregulate during sleep? Experientia ing thresholds in REM and NREM sleep stages. Per- 30:1279-1280,1974. cept Mot Skills 22:927-942, 1966. Shurley JT: Effect of the air-fluidized bed on sleep Rechtschaffen A, Kales A (eds): A Manual of Standard- patterns in healthy human subjects, in Anz CP, Haigert ized Terminology, Techniques and Scoring System for TS (eds): Air-Fluidized Bed,Clinical and Research Sleep Stages of Human Suhjects. Los Angeles, Brain Symposium. Medical University of S. Carolina. Charles- Information Service/Brain Research Institute, UCLA, ton, 1971, pp 38-49. 1968. Simpson RG: Nocturnal disorders of medical interest. Rechtschaffen A, Monroe J:Laboratory studies of in- Practitioner 202:259-268, 1969. somnia, in Kales A (ed): Sleep: Physiology & Pathol- ogy.JB Lippincott Co, Philadelphia, 1969, pp 158-169. Snyder F: Psychophysiology of human sleep. Clin Neurosurg 18: 503-536, 1971. Rechtschaffen A, Wolpert E, Dement W: Nocturnal sleep of narcoleptics. Electroencephalogr Clio Neuro- Starfield B: Enuresis: Its pathogenesis and manage- pbysiol 15:599-609, 1963. ment. Clin Pediatr 11:343-348, 1972.

Reding GR, Zepelin H, Robinson JE Jr, et al: Noc- Steinschneider A: Prolonged apnea and the sudden turnal teeth-grinding:All-night psychophysiologic infant death syndrome: Clinical and laboratory obser- studies. J Dent Res 47: 786-797, 1968. vations. Pediatrics 50:646-654, 1972.

RObinson TM: Presleep activity and sleep quality of Sterman MB: Sleep. in DiCara KV (ed): Limbic and good and poor sleepers. University of Chicago, un- Autonomic Nervous Systems Research.New York, pubhshed doctoral dissertation, June 1969. Plenum Press, 1974, pp 395-417.

75 Sterman MB, Howe RC, Macdonald LR: Facilitation of sleep. Electroencephalogr Clin Neurophysiol 16: of spindle-burst sleep by conditioning of electro- 269-279,1964. encephalographic activity while awake. Science 167: Williams RL, Karacan I: Clinical disorders of sleep, in 1146-1148,1970. Usdi n G (ed): Sleep Research and Clinical Practice. Townsend RE, johnson CC, Muzet A: Effects of long- New York, Brunner/Mazel, 1973, pp 23-58. term exposure [0 tone pulse noise on human sleep. Williams RL, Karacan I: Sleep disorders and dis- Psychophysiology 10:369-376, 1973. ordered sleep, in Ari eti S (ed): American Handbook of Trask CH, Cree EM: Oximeter studies on patients with Psychiatry, Vol 4, Organic Disorders and Psychoso- chronic obstructive emphysema, awake and during matic Medicine. New York, Basic Books, 1975, pp sleep. N Engl J Med 266:639-642, 1962. 854-904.

ValatxjL,Roussel B, Cure M: Sommeil et temperature Williams RL, Karacan I, Hursch CJ: Electroenceph- cerebrate du rat au cours de I'exposition chronique en alography (EEG) of Human Sleep: Clinical Applica- ambiance chaude. Brain Res 55:107-122,1973. tiona. New York, john Wiley and Sons, 1974. Wurtman Rj, Fernstrom JD: Control of brain serotonin Van den Burg W, Van den Hoofdakker RH: Total by the diet. Adv Neurol 5: 19-29, 1974. sleep deprivation on endogenous depression. Arch Cen Psychiatry 32:1121-1125,1975. Wyatt Rj:The serotonin-catecholamine-dream bi- cycle: A clinical study. BioI Psychiatry 5:33-64, 1972. Vogel G, McAbee R, Barker K, et al: REM pressure and improvement of endogenous depression, in Chase Wyatt Rj, Termini BA, Davis J: Biochemical and sleep MH, Mitler MM, Walter PL (eds):Sleep Research, Los studies of schizoph renia - a review of the literatu re, Angeles, Brain Information Service/Brain Research 1960-1970: Part II. Sleep studies. Schizophrenia Bul- Institute, UCLA, 1976, Vol 5, P 15!. letin No.4, 45-66, 1971. Reviewed in Sleep Research 2:378, 1973. Vogel GW, Thurmond A, Gibbons P, et al: Reduction effects on depression syndromes.Arch Gen Psychia- Zarcone V: Narcolepsy.N Engl ] Med 288: 1156- :I try 32:765-777, 1975. 1166, 1973. Wagner M K, Mooney DK: Personality characteristics Zarcone V: REM phase deprivation and schizo- I of long and short sleepers.] Clin Psychol 31 :434-436, phrenia II. Arch Cen Psychiatry 32:1431-1436, 1975. Ii 1975. Zarcone V, Gulevich C, Pivik T, et al: Partial REM Webb WB, Ades H: Sleep tendencies:Effects of baro- phase deprivation and schizophrenia.Arch Gen Psy- metric pressure. Science 143:263-264, 1964. chiatry 18:194-202, 1968. I ) Webb WB, Agnew HW: Are we chronically sleep de- Ziegler AJ: Dream emotions in relation to room tem- prived? Bull Psycho nomic Soc 6:47-48, 1975. perature, in Koella WP, Levin P (eds): Physiology, I Biochemistry, Psychology, Pharmacology, Clinical Wilse WB, Friel j: Sleep stage and personality charac- Applications. First Eu ropean Congress on Sleep Re- teristics of "natural" long and short sleepers. Science search, Basel, 1972. Basel,S Karger, 1973. 171:587-588, 1971. Zimmerman WB: Sleep mentation and auditory awak- Weitzman ED, Czeisler CA, Fusco R, et al: Relation- I ening thresholds. Psychophysiology 6:540-549, 1970. ship of cortisol, REM sleep reduction: Effects on depression syndromes. Arch Cen Psychiatry 32:765-777, Zir CM, Smith RA, Parker DC: Human growth hor- 1975. mone release in sleep: Effect of daytime exercise. J i Clio Endocrinol Metab 32:662-665, 1971. Weitzman ED, Czeisler CA, Fusco R, et al: Relation- Zloty RB, Burdick JA. Adamson jD:Sleep of distance ship of cortisol, growth hormone, body temperature, runners. Activitas Nervosa Superior (PRAHA) 15: i and sleep in man living in an environment free of time 217-221, 1973. cues, in Chase MH, Mitler MM, Walter PL (eds): Sleep Research, Los Angeles, Brain Information Service/ Zubek jP, Welch G, Saunders MG: Electroencephalo- , Brain Research Institute,UCLA, 1976,Vol 5, P 219. graphic changes during and after 14 days of percep- tual deprivation. Science 139:490-492, 1963. Weitzman ED, Goldmacher D, Kripke D, et al:Re- versal of sleep-waking cycle: Effect on sleep stage pattern and certain neuroendocrine rhythms. Trans Am 1 Neurol Assoc 93:153-157, 1968.

Weitzman ED, Graziani L:Sleep and sudden infant death syndrome: A new hypothesis, in Weitzman ED (ed): Advances in Sleep Research, Flushing,New York, Spectrum Publications, Vol I, 1974, pp 327- 341. I Wever R: ELF-effects on human circadian rhythms, t in Persinger MA (ed):ELF and VLFElectromagnetic Field Effects, New York, Plenum, 1974, pp 101-144. j Wilkinson RT: Loss of sleep. Proc R Soc Med 62:27- 28,1969. I Williams HL, Hammack jT, Daly RC, et al: Responses I to auditory stimulation, sleep loss and the EEG stages 76 I j ;